Problem 1¶
Imports¶
In [ ]:
import numpy as np
import pandas as pd
from tensorflow import keras
import tensorflow as tf
from tensorflow.keras import layers
from matplotlib import pyplot as plt
from keras import backend as kb
import matplotlib.pyplot as plt
from sklearn.manifold import TSNE
import seaborn as sns
%matplotlib inline
Load Data into Numpy Arrays¶
In [ ]:
#Using Keras load_data extract values for mnist into train and test values
#(X_train, y_train), (X_test, y_test) = keras.datasets.mnist.load_data()
#Using Keras load_data extract values for mnist into train and test values
(X_train, y_train), (X_test, y_test) = keras.datasets.mnist.load_data()
X_train, X_test = X_train/255.0, X_test/255.0
X_train = X_train.astype("float32")
X_test = X_test.astype("float32")
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz 11493376/11490434 [==============================] - 0s 0us/step
In [ ]:
#What does the data look like?
print(X_train.shape)
print(y_train.shape)
print(X_test.shape)
print(y_test.shape)
(60000, 28, 28) (60000,) (10000, 28, 28) (10000,)
In [ ]:
# What does a sample image look like? This is the 10,000th indexed value. This looks like 3.
plt.imshow(X_train[10000], cmap='gray')
plt.show()
In [ ]:
# Is the label activated for 10,000 index to show 3?
y_train[10000]
Out[ ]:
3
The label shows 3, but this is a problem. We need to have categorical values of 0's and 1's as oposed to the actual labels.
In [ ]:
#Use Keras utility to transform labels into onehot encoding (binary categories)
y_train = keras.utils.to_categorical(y_train, 10)
y_test = keras.utils.to_categorical(y_test, 10)
In [ ]:
# This looks like a three, lets check label for the 10,000th index value. Looks like the 3rd value (4th index) was activated.
y_train[10000]
Out[ ]:
array([0., 0., 0., 1., 0., 0., 0., 0., 0., 0.], dtype=float32)
Create DL Model¶
In [ ]:
shape = (28, 28, 1) # Define shape of input for Keras model
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(1024,activation='relu'),
tf.keras.layers.Dense(1024,activation='relu'),
tf.keras.layers.Dense(1024,activation='relu'),
tf.keras.layers.Dense(1024,activation='relu'),
tf.keras.layers.Dense(1024,activation='relu'),
tf.keras.layers.Dense(10, activation='softmax')
]
)
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten (Flatten) (None, 784) 0 _________________________________________________________________ dense (Dense) (None, 1024) 803840 _________________________________________________________________ dense_1 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_2 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_3 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_4 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_5 (Dense) (None, 10) 10250 ================================================================= Total params: 5,012,490 Trainable params: 5,012,490 Non-trainable params: 0 _________________________________________________________________
Compile and Fit DL Model¶
In [ ]:
#Define model
model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=["accuracy"])
#Fit model
model.fit(X_train, y_train, batch_size=64, epochs=10, validation_split=0.1)
Epoch 1/10 844/844 [==============================] - 3s 4ms/step - loss: 0.2430 - accuracy: 0.9309 - val_loss: 0.1202 - val_accuracy: 0.9655 Epoch 2/10 844/844 [==============================] - 3s 4ms/step - loss: 0.1177 - accuracy: 0.9682 - val_loss: 0.0974 - val_accuracy: 0.9738 Epoch 3/10 844/844 [==============================] - 3s 4ms/step - loss: 0.0901 - accuracy: 0.9757 - val_loss: 0.0945 - val_accuracy: 0.9753 Epoch 4/10 844/844 [==============================] - 3s 4ms/step - loss: 0.0732 - accuracy: 0.9810 - val_loss: 0.0824 - val_accuracy: 0.9777 Epoch 5/10 844/844 [==============================] - 3s 4ms/step - loss: 0.0599 - accuracy: 0.9836 - val_loss: 0.1060 - val_accuracy: 0.9755 Epoch 6/10 844/844 [==============================] - 3s 3ms/step - loss: 0.0561 - accuracy: 0.9855 - val_loss: 0.1002 - val_accuracy: 0.9757 Epoch 7/10 844/844 [==============================] - 3s 4ms/step - loss: 0.0416 - accuracy: 0.9890 - val_loss: 0.0949 - val_accuracy: 0.9800 Epoch 8/10 844/844 [==============================] - 3s 4ms/step - loss: 0.0374 - accuracy: 0.9898 - val_loss: 0.0728 - val_accuracy: 0.9808 Epoch 9/10 844/844 [==============================] - 3s 4ms/step - loss: 0.0348 - accuracy: 0.9904 - val_loss: 0.1107 - val_accuracy: 0.9767 Epoch 10/10 844/844 [==============================] - 3s 3ms/step - loss: 0.0354 - accuracy: 0.9910 - val_loss: 0.1072 - val_accuracy: 0.9790
Out[ ]:
<tensorflow.python.keras.callbacks.History at 0x7f26a00f7ef0>
Evaluate Test Data¶
In [ ]:
score = model.evaluate(X_test, y_test, verbose=1)
313/313 [==============================] - 1s 2ms/step - loss: 0.1078 - accuracy: 0.9811
Create Function to extract layer data¶
In [ ]:
def extract_layer_output(data, layer_index):
layer_output = []
keras_function = kb.function([model.input], [model.get_layer(index=layer_index).output])
layer_output.append(keras_function([data, 1]))
layer_output = np.squeeze(layer_output) # remove all dimensions of size 1 i.e., (1,1,10000,1024) to (10000, 1024)
return(layer_output)
Test Data Analysis¶
In [ ]:
#Get 1000 samples from test data
X_test_1000 = X_test[0:1000]
y_test_1000 = y_test[0:1000]
Softmax Predictions¶
In [ ]:
#Get softmax layer output for 1000 test samples
softmax_output = extract_layer_output(X_test_1000, 6) #6 is last softmax layer
#Get max index for each matrix (convert from matrix of 10 labels to 1 which is the prediction)
softmax_output = np.argmax(softmax_output,axis=1)
In [ ]:
softmax_output.shape
Out[ ]:
(1000,)
In [ ]:
#Get the first 10 predictions for each nbr from 0-10. First 10 are 0, then 1, etc..
l = []
for x in range(0,10):
t = np.where(softmax_output==x)
t = np.array(t)
t = np.squeeze(t)
t = t[0:10]
l.append(t)
l = np.array(l)
l = l.reshape(100)
In [ ]:
print(l)
[ 3 10 13 25 28 55 69 71 101 126 2 5 14 29 31 37 39 40 46 57 1 35 38 43 47 72 77 82 106 119 18 30 32 44 51 63 68 76 87 90 4 6 19 24 27 33 42 48 49 56 8 15 23 45 52 53 59 102 120 127 11 21 22 50 54 66 81 88 91 98 0 17 26 34 36 41 60 64 70 75 61 84 110 128 134 146 177 179 181 184 7 9 12 16 20 58 62 73 78 92]
In [ ]:
#Create 10X10 plot that shows predictions for first 10 of each nbr. If predictions are good all nbrs should match on each row
num_row = 10
num_col = 10
fig = plt.figure
fig, axes = plt.subplots(num_row, num_col, figsize=(1.5*num_col,2*num_row))
for i,tt in enumerate(l):
ax = axes[i//num_col, i%num_col]
ax.imshow(X_test_1000[tt], cmap='gray')
ax.set_xticks([])
ax.set_yticks([])
plt.show()
Second to Last Layer Predictions¶
In [ ]:
#Get softmax layer output for 1000 test samples
secondlast = extract_layer_output(X_test_1000, 5) #5 is the second to last layer (last hidden)
#Get max index for each matrix (convert from matrix of 10 labels to 1 which is the prediction)
#secondlast = extract_layer_output(X_test_1000, 5) #6 is current last softmax layer
In [ ]:
#Select 10 random int's
int = np.random.randint(0,secondlast.shape[1],10)
int = np.sort(int)
In [ ]:
secondlast = secondlast[:,int]
In [ ]:
#Get max index for each matrix (convert from matrix of 10 labels to 1 which is the prediction)
secondlast = np.argmax(secondlast,axis=1)
In [ ]:
#Get the first 10 predictions for each nbr from 0-10. First 10 are 0, then 1, etc..
l = []
for x in range(0,10):
t = np.where(secondlast==x)
t = np.array(t)
if t.size < 10:
t = np.full(10,-1)
t = np.array(t)
t = np.squeeze(t)
t = t[0:10]
l.append(t)
l = np.array(l)
l = l.reshape(100)
In [ ]:
#Create 10X10 plot that shows predictions for first 10 of each nbr. If predictions are good all nbrs should match on each row
num_row = 10
num_col = 10
fig = plt.figure
fig, axes = plt.subplots(num_row, num_col, figsize=(1.5*num_col,2*num_row))
for i,tt in enumerate(l):
ax = axes[i//num_col, i%num_col]
#print(tt)
if tt==-1:
''
else:
ax.imshow(X_test_1000[tt], cmap='gray')
ax.set_xticks([])
ax.set_yticks([])
plt.show()
Analysis¶
These results were confusing to me and really dont make sense. Since the 10 of the 1024 dimensions are chosen at random there is no way to tie them back to actual labels. The results are very random and do not help explain relationships in the data compared to the actual labels. It does appear to somewhat group results, but is not clear to me how this would be helpful.
Apply TSNE to reduce and visualize dimensions¶
In [ ]:
#Create funtion to generate DF's for dimensions and means for visualization
def createDF(data, labels):
df_test = pd.DataFrame({'x': data[:, 0], 'y': data[:, 1], 'label': labels})
df_mean = df_test.groupby('label').mean()
return df_test, df_mean
In [ ]:
#Convert test labels to true values by index
test_labels = np.argmax(y_test_1000,axis=1)
Raw Data - 1000 examples¶
In [ ]:
#Apply TSNE to test data
layer_data = TSNE(n_components=2).fit_transform(X_test_1000.reshape(1000,784))
#Call function to get DF and DF with mean values for visualization
df_test, df_mean = createDF(layer_data, test_labels)
In [ ]:
#Plot 1000 test examples using TSNE dimension reduction
p1 = sns.lmplot(data=df_test, x="x", y="y", fit_reg=False, hue="label")
for index, row in df_mean.iterrows():
plt.annotate(index, xy=(row[0],row[1]), weight='bold', fontsize=20)
#sns.plt.show()
First Layer - 1000 examples¶
In [ ]:
#Get first layer output for 1000 test samples
layer_data = extract_layer_output(X_test_1000, 1)
#Apply TSNE to test data
layer_data = TSNE(n_components=2).fit_transform(layer_data.reshape(1000,1024))
#Call function to get DF and DF with mean values for visualization
df_test, df_mean = createDF(layer_data, test_labels)
In [ ]:
#Plot 1000 test examples using TSNE dimension reduction
p1 = sns.lmplot(data=df_test, x="x", y="y", fit_reg=False, hue="label")
for index, row in df_mean.iterrows():
plt.annotate(index, xy=(row[0],row[1]), weight='bold', fontsize=20)
#sns.plt.show()
2nd Layer - 1000 examples¶
In [ ]:
#Get second layer output for 1000 test samples
layer_data = extract_layer_output(X_test_1000, 2)
#Apply TSNE to test data
layer_data = TSNE(n_components=2).fit_transform(layer_data.reshape(1000,1024))
#Call function to get DF and DF with mean values for visualization
df_test, df_mean = createDF(layer_data, test_labels)
In [ ]:
#Plot 1000 test examples using TSNE dimension reduction
p1 = sns.lmplot(data=df_test, x="x", y="y", fit_reg=False, hue="label")
for index, row in df_mean.iterrows():
plt.annotate(index, xy=(row[0],row[1]), weight='bold', fontsize=20)
#sns.plt.show()
3rd Layer - 1000 examples¶
In [ ]:
#Get third layer output for 1000 test samples
layer_data = extract_layer_output(X_test_1000, 3)
#Apply TSNE to test data
layer_data = TSNE(n_components=2).fit_transform(layer_data.reshape(1000,1024))
#Call function to get DF and DF with mean values for visualization
df_test, df_mean = createDF(layer_data, test_labels)
In [ ]:
#Plot 1000 test examples using TSNE dimension reduction
p1 = sns.lmplot(data=df_test, x="x", y="y", fit_reg=False, hue="label")
for index, row in df_mean.iterrows():
plt.annotate(index, xy=(row[0],row[1]), weight='bold', fontsize=20)
#sns.plt.show()
4th Layer - 1000 examples¶
In [ ]:
#Get fourth layer output for 1000 test samples
layer_data = extract_layer_output(X_test_1000, 4)
#Apply TSNE to test data
layer_data = TSNE(n_components=2).fit_transform(layer_data.reshape(1000,1024))
#Call function to get DF and DF with mean values for visualization
df_test, df_mean = createDF(layer_data, test_labels)
In [ ]:
#Plot 1000 test examples using TSNE dimension reduction
p1 = sns.lmplot(data=df_test, x="x", y="y", fit_reg=False, hue="label")
for index, row in df_mean.iterrows():
plt.annotate(index, xy=(row[0],row[1]), weight='bold', fontsize=20)
#sns.plt.show()
5th Layer - 1000 examples¶
In [ ]:
#Get fifth layer output for 1000 test samples
layer_data = extract_layer_output(X_test_1000, 5)
#Apply TSNE to test data
layer_data = TSNE(n_components=2).fit_transform(layer_data.reshape(1000,1024))
#Call function to get DF and DF with mean values for visualization
df_test, df_mean = createDF(layer_data, test_labels)
In [ ]:
#Plot 1000 test examples using TSNE dimension reduction
p1 = sns.lmplot(data=df_test, x="x", y="y", fit_reg=False, hue="label")
for index, row in df_mean.iterrows():
plt.annotate(index, xy=(row[0],row[1]), weight='bold', fontsize=20)
#sns.plt.show()
Last Layer (softmax) - 1000 examples¶
In [ ]:
#Get last layer output for 1000 test samples
layer_data = extract_layer_output(X_test_1000, 6)
#Apply TSNE to test data
layer_data = TSNE(n_components=2).fit_transform(layer_data.reshape(1000,10))
#Call function to get DF and DF with mean values for visualization
df_test, df_mean = createDF(layer_data, test_labels)
In [ ]:
#Plot 1000 test examples using TSNE dimension reduction
p1 = sns.lmplot(data=df_test, x="x", y="y", fit_reg=False, hue="label")
for index, row in df_mean.iterrows():
plt.annotate(index, xy=(row[0],row[1]), weight='bold', fontsize=20)
#sns.plt.show()
Analysis¶
I tried both PCA and TSNE and TSNE results were much easier for me to undestand. As I reduced the dimensions for each layer the groupings get closer and closer as you make it to the last layer (softmax output). Using dimension reduction does seem very helpful for understandnig the population of the data.
Problem 2¶
Imports¶
In [ ]:
import numpy as np
from tensorflow import keras
import tensorflow as tf
from tensorflow.keras import layers
import scipy
!pip install librosa # in colab, you’ll need to install this
import librosa
Requirement already satisfied: librosa in /usr/local/lib/python3.6/dist-packages (0.6.3) Requirement already satisfied: resampy>=0.2.0 in /usr/local/lib/python3.6/dist-packages (from librosa) (0.2.2) Requirement already satisfied: scipy>=1.0.0 in /usr/local/lib/python3.6/dist-packages (from librosa) (1.4.1) Requirement already satisfied: scikit-learn!=0.19.0,>=0.14.0 in /usr/local/lib/python3.6/dist-packages (from librosa) (0.22.2.post1) Requirement already satisfied: numba>=0.38.0 in /usr/local/lib/python3.6/dist-packages (from librosa) (0.48.0) Requirement already satisfied: numpy>=1.8.0 in /usr/local/lib/python3.6/dist-packages (from librosa) (1.18.5) Requirement already satisfied: six>=1.3 in /usr/local/lib/python3.6/dist-packages (from librosa) (1.15.0) Requirement already satisfied: audioread>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from librosa) (2.1.8) Requirement already satisfied: decorator>=3.0.0 in /usr/local/lib/python3.6/dist-packages (from librosa) (4.4.2) Requirement already satisfied: joblib>=0.12 in /usr/local/lib/python3.6/dist-packages (from librosa) (0.16.0) Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from numba>=0.38.0->librosa) (50.3.0) Requirement already satisfied: llvmlite<0.32.0,>=0.31.0dev0 in /usr/local/lib/python3.6/dist-packages (from numba>=0.38.0->librosa) (0.31.0)
In [ ]:
def readAudio(file_name):
s, sr = librosa.load(file_name, sr=None)
S = librosa.stft(s, n_fft=1024, hop_length=512)
S_abs = np.abs(S).T #get absolute values and transpose
S = S.T #transpose values
return(S, S_abs, sr)
In [ ]:
#Read main and test audio files and convert to spectrograms
X_train, X_train_abs, sr = readAudio('train_clean_male.wav')
y_train, y_train_abs, sr = readAudio('train_dirty_male.wav')
test1, test1_abs, sr = readAudio('test_x_01.wav')
test2, test2_abs, sr = readAudio('test_x_02.wav')
In [ ]:
#Read in main audio files and convert to spectrograms
#s, sr = librosa.load('train_clean_male.wav', sr=None)
#S = librosa.stft(s, n_fft=1024, hop_length=512)
#sn, sr = librosa.load('train_dirty_male.wav', sr=None)
#X = librosa.stft(sn, n_fft=1024, hop_length=512)
#Read in test audio files and convert to spectrograms
#X, sr = librosa.load('test_x_01.wav', sr=None)
#X_test = librosa.stft(X, n_fft=1024, hop_length=512)
#X, sr = librosa.load('test_x_02.wav', sr=None)
#X_test2 = librosa.stft(X, n_fft=1024, hop_length=512)
In [ ]:
#S = np.abs(S)
#S=S.T
#X = np.abs(X)
#X=X.T
In [ ]:
#Create DNN model with 2 hidden layers
shape = (2459, 513) # Define shape of input for Keras model
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Dense(513,activation='relu'),
tf.keras.layers.Dense(513,activation='relu'),
tf.keras.layers.Dense(513, activation='relu')
]
)
model.summary()
Model: "sequential_2" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= dense_9 (Dense) (None, 2459, 513) 263682 _________________________________________________________________ dense_10 (Dense) (None, 2459, 513) 263682 _________________________________________________________________ dense_11 (Dense) (None, 2459, 513) 263682 ================================================================= Total params: 791,046 Trainable params: 791,046 Non-trainable params: 0 _________________________________________________________________
In [ ]:
#Compile and fit the model
model.compile(loss="MeanSquaredError", optimizer="adam", metrics=["accuracy"])
model.fit(X_train_abs, y_train_abs, batch_size=50, epochs=50, validation_split=0.1)
Epoch 1/50
WARNING:tensorflow:Model was constructed with shape (None, 2459, 513) for input Tensor("input_3:0", shape=(None, 2459, 513), dtype=float32), but it was called on an input with incompatible shape (None, 513).
WARNING:tensorflow:Model was constructed with shape (None, 2459, 513) for input Tensor("input_3:0", shape=(None, 2459, 513), dtype=float32), but it was called on an input with incompatible shape (None, 513).
43/45 [===========================>..] - ETA: 0s - loss: 0.0662 - accuracy: 0.1944WARNING:tensorflow:Model was constructed with shape (None, 2459, 513) for input Tensor("input_3:0", shape=(None, 2459, 513), dtype=float32), but it was called on an input with incompatible shape (None, 513).
45/45 [==============================] - 0s 5ms/step - loss: 0.0659 - accuracy: 0.2015 - val_loss: 0.0425 - val_accuracy: 0.3537
Epoch 2/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0373 - accuracy: 0.4076 - val_loss: 0.0380 - val_accuracy: 0.3740
Epoch 3/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0316 - accuracy: 0.4510 - val_loss: 0.0359 - val_accuracy: 0.4553
Epoch 4/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0280 - accuracy: 0.5011 - val_loss: 0.0345 - val_accuracy: 0.4593
Epoch 5/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0262 - accuracy: 0.5011 - val_loss: 0.0338 - val_accuracy: 0.4675
Epoch 6/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0247 - accuracy: 0.5368 - val_loss: 0.0326 - val_accuracy: 0.4878
Epoch 7/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0238 - accuracy: 0.5404 - val_loss: 0.0333 - val_accuracy: 0.4919
Epoch 8/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0228 - accuracy: 0.5490 - val_loss: 0.0327 - val_accuracy: 0.4715
Epoch 9/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0222 - accuracy: 0.5554 - val_loss: 0.0328 - val_accuracy: 0.4797
Epoch 10/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0217 - accuracy: 0.5630 - val_loss: 0.0331 - val_accuracy: 0.4756
Epoch 11/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0213 - accuracy: 0.5635 - val_loss: 0.0329 - val_accuracy: 0.4431
Epoch 12/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0211 - accuracy: 0.5599 - val_loss: 0.0332 - val_accuracy: 0.4797
Epoch 13/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0206 - accuracy: 0.5477 - val_loss: 0.0324 - val_accuracy: 0.4675
Epoch 14/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0199 - accuracy: 0.5626 - val_loss: 0.0331 - val_accuracy: 0.5000
Epoch 15/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0199 - accuracy: 0.5540 - val_loss: 0.0330 - val_accuracy: 0.4797
Epoch 16/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0190 - accuracy: 0.5612 - val_loss: 0.0334 - val_accuracy: 0.4837
Epoch 17/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0185 - accuracy: 0.5662 - val_loss: 0.0330 - val_accuracy: 0.4837
Epoch 18/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0173 - accuracy: 0.5635 - val_loss: 0.0329 - val_accuracy: 0.4878
Epoch 19/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0172 - accuracy: 0.5725 - val_loss: 0.0334 - val_accuracy: 0.4797
Epoch 20/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0165 - accuracy: 0.5716 - val_loss: 0.0326 - val_accuracy: 0.4797
Epoch 21/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0164 - accuracy: 0.5703 - val_loss: 0.0329 - val_accuracy: 0.4959
Epoch 22/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0160 - accuracy: 0.5703 - val_loss: 0.0331 - val_accuracy: 0.4959
Epoch 23/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0158 - accuracy: 0.5689 - val_loss: 0.0329 - val_accuracy: 0.4878
Epoch 24/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0154 - accuracy: 0.5703 - val_loss: 0.0330 - val_accuracy: 0.4959
Epoch 25/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0151 - accuracy: 0.5725 - val_loss: 0.0328 - val_accuracy: 0.5081
Epoch 26/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0144 - accuracy: 0.5784 - val_loss: 0.0334 - val_accuracy: 0.4837
Epoch 27/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0140 - accuracy: 0.5834 - val_loss: 0.0336 - val_accuracy: 0.4878
Epoch 28/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0140 - accuracy: 0.5798 - val_loss: 0.0328 - val_accuracy: 0.4837
Epoch 29/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0131 - accuracy: 0.5825 - val_loss: 0.0337 - val_accuracy: 0.4878
Epoch 30/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0130 - accuracy: 0.5883 - val_loss: 0.0342 - val_accuracy: 0.5000
Epoch 31/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0127 - accuracy: 0.5838 - val_loss: 0.0338 - val_accuracy: 0.4919
Epoch 32/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0123 - accuracy: 0.5888 - val_loss: 0.0334 - val_accuracy: 0.4797
Epoch 33/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0120 - accuracy: 0.5816 - val_loss: 0.0335 - val_accuracy: 0.5000
Epoch 34/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0122 - accuracy: 0.5911 - val_loss: 0.0334 - val_accuracy: 0.4797
Epoch 35/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0118 - accuracy: 0.5811 - val_loss: 0.0338 - val_accuracy: 0.5041
Epoch 36/50
45/45 [==============================] - 0s 4ms/step - loss: 0.0117 - accuracy: 0.5897 - val_loss: 0.0337 - val_accuracy: 0.5081
Epoch 37/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0133 - accuracy: 0.5861 - val_loss: 0.0332 - val_accuracy: 0.4919
Epoch 38/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0121 - accuracy: 0.5874 - val_loss: 0.0339 - val_accuracy: 0.5041
Epoch 39/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0122 - accuracy: 0.5879 - val_loss: 0.0336 - val_accuracy: 0.4837
Epoch 40/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0109 - accuracy: 0.5865 - val_loss: 0.0335 - val_accuracy: 0.5041
Epoch 41/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0105 - accuracy: 0.5947 - val_loss: 0.0332 - val_accuracy: 0.5000
Epoch 42/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0100 - accuracy: 0.6001 - val_loss: 0.0338 - val_accuracy: 0.4837
Epoch 43/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0098 - accuracy: 0.6019 - val_loss: 0.0334 - val_accuracy: 0.5000
Epoch 44/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0095 - accuracy: 0.6028 - val_loss: 0.0337 - val_accuracy: 0.5041
Epoch 45/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0108 - accuracy: 0.5987 - val_loss: 0.0329 - val_accuracy: 0.4959
Epoch 46/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0103 - accuracy: 0.6042 - val_loss: 0.0337 - val_accuracy: 0.5163
Epoch 47/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0100 - accuracy: 0.6033 - val_loss: 0.0331 - val_accuracy: 0.5163
Epoch 48/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0092 - accuracy: 0.6078 - val_loss: 0.0335 - val_accuracy: 0.5000
Epoch 49/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0095 - accuracy: 0.6100 - val_loss: 0.0335 - val_accuracy: 0.5163
Epoch 50/50
45/45 [==============================] - 0s 3ms/step - loss: 0.0093 - accuracy: 0.6073 - val_loss: 0.0334 - val_accuracy: 0.5244
Out[ ]:
<tensorflow.python.keras.callbacks.History at 0x7f263039ad30>
In [ ]:
#Read in test audio files and convert to spectrograms
#test1, sr=librosa.load('test_x_01.wav', sr=None)
#TEST1=librosa.stft(test1, n_fft=1024, hop_length=512)
#test2, sr=librosa.load('test_x_02.wav', sr=None)
#TEST2=librosa.stft(test2, n_fft=1024, hop_length=512)
In [ ]:
#Get absolute values
#TEST1_abs = np.abs(TEST1)
#TEST2_abs = np.abs(TEST2)
In [ ]:
#Use trained models to perform predictions
test1_predict = model.predict(test1_abs)
test2_predict = model.predict(test2_abs)
WARNING:tensorflow:Model was constructed with shape (None, 2459, 513) for input Tensor("input_3:0", shape=(None, 2459, 513), dtype=float32), but it was called on an input with incompatible shape (None, 513).
In [ ]:
s1 = np.multiply((test1/test1_abs).T, np.abs(test1_predict).T)
s2 = np.multiply((test2/test2_abs).T, np.abs(test2_predict).T)
In [ ]:
scipy.signal.istft(s1)
sh_test1 = scipy.signal.istft(s1)
scipy.signal.istft(s2)
sh_test2 = scipy.signal.istft(s2)
In [ ]:
#Conver to Numpy Arrays
sh_test1 = np.array(sh_test1)
sh_test2 = np.array(sh_test2)
In [ ]:
librosa.output.write_wav('test_s_01_recons.wav', sh_test1, sr)
librosa.output.write_wav('test_s_02_recons.wav', sh_test2, sr)
In [204]:
from IPython.display import Audio
Audio('test_s_01_recons.wav')
Out[204]:
In [205]:
from IPython.display import Audio
Audio('test_s_02_recons.wav')
Out[205]:
Problem 3¶
SGD¶
Imports¶
In [ ]:
import numpy as np
from tensorflow import keras
import tensorflow as tf
from tensorflow.keras import layers
import matplotlib.pyplot as plt
In [ ]:
#Using Keras load_data extract values for mnist into train and test values
(X_train, y_train), (X_test, y_test) = keras.datasets.mnist.load_data()
X_train, X_test = X_train/255.0, X_test/255.0
X_train = X_train.astype("float32")
X_test = X_test.astype("float32")
In [ ]:
#Use Keras utility to transform labels into onehot encoding (binary categories)
y_train = keras.utils.to_categorical(y_train, 10)
y_test = keras.utils.to_categorical(y_test, 10)
(a) Activation function: the logistic sigmoid function; initialization: random numbers gen-erated from the normal distribution (μ = 0, σ = 0.01)¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.01, seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_8" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_8 (Flatten) (None, 784) 0 _________________________________________________________________ dense_48 (Dense) (None, 512) 401920 _________________________________________________________________ dense_49 (Dense) (None, 512) 262656 _________________________________________________________________ dense_50 (Dense) (None, 512) 262656 _________________________________________________________________ dense_51 (Dense) (None, 512) 262656 _________________________________________________________________ dense_52 (Dense) (None, 512) 262656 _________________________________________________________________ dense_53 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.SGD(learning_rate=.01)
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
history = model.fit(X_train, y_train, batch_size=64, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3060 - accuracy: 0.1043 - val_loss: 2.3111 - val_accuracy: 0.1010 Epoch 2/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3059 - accuracy: 0.1059 - val_loss: 2.3042 - val_accuracy: 0.1010 Epoch 3/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3060 - accuracy: 0.1039 - val_loss: 2.3039 - val_accuracy: 0.1028 Epoch 4/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3060 - accuracy: 0.1054 - val_loss: 2.3080 - val_accuracy: 0.1028 Epoch 5/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3057 - accuracy: 0.1055 - val_loss: 2.3075 - val_accuracy: 0.1009 Epoch 6/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3058 - accuracy: 0.1063 - val_loss: 2.3084 - val_accuracy: 0.1135 Epoch 7/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3058 - accuracy: 0.1058 - val_loss: 2.3072 - val_accuracy: 0.1009 Epoch 8/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3059 - accuracy: 0.1054 - val_loss: 2.3050 - val_accuracy: 0.1135 Epoch 9/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3054 - accuracy: 0.1075 - val_loss: 2.3055 - val_accuracy: 0.1135 Epoch 10/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3057 - accuracy: 0.1062 - val_loss: 2.3052 - val_accuracy: 0.1010 Epoch 11/200 938/938 [==============================] - 3s 4ms/step - loss: 2.3053 - accuracy: 0.1053 - val_loss: 2.3047 - val_accuracy: 0.0980 Epoch 12/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3055 - accuracy: 0.1047 - val_loss: 2.3040 - val_accuracy: 0.1135 Epoch 13/200 938/938 [==============================] - 3s 4ms/step - loss: 2.3057 - accuracy: 0.1041 - val_loss: 2.3026 - val_accuracy: 0.1135 Epoch 14/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3050 - accuracy: 0.1061 - val_loss: 2.3061 - val_accuracy: 0.0982 Epoch 15/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3052 - accuracy: 0.1051 - val_loss: 2.3062 - val_accuracy: 0.1135 Epoch 16/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3053 - accuracy: 0.1050 - val_loss: 2.3028 - val_accuracy: 0.1135 Epoch 17/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3053 - accuracy: 0.1061 - val_loss: 2.3082 - val_accuracy: 0.1010 Epoch 18/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3055 - accuracy: 0.1037 - val_loss: 2.3059 - val_accuracy: 0.1009 Epoch 19/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3052 - accuracy: 0.1045 - val_loss: 2.3096 - val_accuracy: 0.0958 Epoch 20/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3052 - accuracy: 0.1054 - val_loss: 2.3035 - val_accuracy: 0.1028 Epoch 21/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3054 - accuracy: 0.1051 - val_loss: 2.3065 - val_accuracy: 0.1135 Epoch 22/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3048 - accuracy: 0.1081 - val_loss: 2.3052 - val_accuracy: 0.1135 Epoch 23/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3049 - accuracy: 0.1064 - val_loss: 2.3038 - val_accuracy: 0.0974 Epoch 24/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3049 - accuracy: 0.1050 - val_loss: 2.3060 - val_accuracy: 0.1028 Epoch 25/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3049 - accuracy: 0.1038 - val_loss: 2.3052 - val_accuracy: 0.1135 Epoch 26/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3045 - accuracy: 0.1087 - val_loss: 2.3053 - val_accuracy: 0.1010 Epoch 27/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3047 - accuracy: 0.1058 - val_loss: 2.3025 - val_accuracy: 0.1135 Epoch 28/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3050 - accuracy: 0.1044 - val_loss: 2.3044 - val_accuracy: 0.1135 Epoch 29/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3047 - accuracy: 0.1062 - val_loss: 2.3064 - val_accuracy: 0.0980 Epoch 30/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3046 - accuracy: 0.1054 - val_loss: 2.3052 - val_accuracy: 0.0974 Epoch 31/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3046 - accuracy: 0.1064 - val_loss: 2.3053 - val_accuracy: 0.0982 Epoch 32/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3049 - accuracy: 0.1058 - val_loss: 2.3039 - val_accuracy: 0.0980 Epoch 33/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3044 - accuracy: 0.1069 - val_loss: 2.3069 - val_accuracy: 0.1028 Epoch 34/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3045 - accuracy: 0.1078 - val_loss: 2.3046 - val_accuracy: 0.1032 Epoch 35/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3045 - accuracy: 0.1059 - val_loss: 2.3059 - val_accuracy: 0.1135 Epoch 36/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3043 - accuracy: 0.1074 - val_loss: 2.3032 - val_accuracy: 0.0958 Epoch 37/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3046 - accuracy: 0.1051 - val_loss: 2.3037 - val_accuracy: 0.1028 Epoch 38/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3044 - accuracy: 0.1076 - val_loss: 2.3044 - val_accuracy: 0.1135 Epoch 39/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3043 - accuracy: 0.1069 - val_loss: 2.3055 - val_accuracy: 0.0892 Epoch 40/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3046 - accuracy: 0.1037 - val_loss: 2.3023 - val_accuracy: 0.1032 Epoch 41/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3042 - accuracy: 0.1086 - val_loss: 2.3036 - val_accuracy: 0.0982 Epoch 42/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3041 - accuracy: 0.1060 - val_loss: 2.3044 - val_accuracy: 0.1135 Epoch 43/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3040 - accuracy: 0.1073 - val_loss: 2.3053 - val_accuracy: 0.0958 Epoch 44/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3043 - accuracy: 0.1064 - val_loss: 2.3025 - val_accuracy: 0.1135 Epoch 45/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3042 - accuracy: 0.1052 - val_loss: 2.3036 - val_accuracy: 0.0980 Epoch 46/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3041 - accuracy: 0.1082 - val_loss: 2.3059 - val_accuracy: 0.1135 Epoch 47/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3041 - accuracy: 0.1065 - val_loss: 2.3038 - val_accuracy: 0.1010 Epoch 48/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3037 - accuracy: 0.1086 - val_loss: 2.3029 - val_accuracy: 0.1028 Epoch 49/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3039 - accuracy: 0.1067 - val_loss: 2.3030 - val_accuracy: 0.1135 Epoch 50/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3040 - accuracy: 0.1064 - val_loss: 2.3032 - val_accuracy: 0.1010 Epoch 51/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3042 - accuracy: 0.1076 - val_loss: 2.3035 - val_accuracy: 0.1135 Epoch 52/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3039 - accuracy: 0.1070 - val_loss: 2.3019 - val_accuracy: 0.0980 Epoch 53/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3039 - accuracy: 0.1067 - val_loss: 2.3033 - val_accuracy: 0.1135 Epoch 54/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3037 - accuracy: 0.1079 - val_loss: 2.3030 - val_accuracy: 0.1135 Epoch 55/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3037 - accuracy: 0.1068 - val_loss: 2.3046 - val_accuracy: 0.1135 Epoch 56/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3039 - accuracy: 0.1077 - val_loss: 2.3030 - val_accuracy: 0.1135 Epoch 57/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3040 - accuracy: 0.1072 - val_loss: 2.3022 - val_accuracy: 0.1135 Epoch 58/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1080 - val_loss: 2.3075 - val_accuracy: 0.0980 Epoch 59/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3038 - accuracy: 0.1052 - val_loss: 2.3056 - val_accuracy: 0.1010 Epoch 60/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1071 - val_loss: 2.3031 - val_accuracy: 0.1028 Epoch 61/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1067 - val_loss: 2.3043 - val_accuracy: 0.0974 Epoch 62/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3037 - accuracy: 0.1065 - val_loss: 2.3032 - val_accuracy: 0.1028 Epoch 63/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1081 - val_loss: 2.3027 - val_accuracy: 0.0980 Epoch 64/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3037 - accuracy: 0.1066 - val_loss: 2.3040 - val_accuracy: 0.1135 Epoch 65/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3035 - accuracy: 0.1071 - val_loss: 2.3032 - val_accuracy: 0.1135 Epoch 66/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1070 - val_loss: 2.3028 - val_accuracy: 0.1135 Epoch 67/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1087 - val_loss: 2.3039 - val_accuracy: 0.1135 Epoch 68/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1088 - val_loss: 2.3036 - val_accuracy: 0.1135 Epoch 69/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1063 - val_loss: 2.3023 - val_accuracy: 0.1135 Epoch 70/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1063 - val_loss: 2.3033 - val_accuracy: 0.1135 Epoch 71/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3035 - accuracy: 0.1073 - val_loss: 2.3026 - val_accuracy: 0.1010 Epoch 72/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3034 - accuracy: 0.1077 - val_loss: 2.3040 - val_accuracy: 0.1028 Epoch 73/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3032 - accuracy: 0.1073 - val_loss: 2.3045 - val_accuracy: 0.1010 Epoch 74/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3032 - accuracy: 0.1100 - val_loss: 2.3031 - val_accuracy: 0.1135 Epoch 75/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3036 - accuracy: 0.1056 - val_loss: 2.3026 - val_accuracy: 0.0980 Epoch 76/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3031 - accuracy: 0.1082 - val_loss: 2.3047 - val_accuracy: 0.0980 Epoch 77/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3032 - accuracy: 0.1075 - val_loss: 2.3028 - val_accuracy: 0.1010 Epoch 78/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3035 - accuracy: 0.1066 - val_loss: 2.3022 - val_accuracy: 0.1135 Epoch 79/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3033 - accuracy: 0.1073 - val_loss: 2.3028 - val_accuracy: 0.0958 Epoch 80/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3031 - accuracy: 0.1069 - val_loss: 2.3040 - val_accuracy: 0.0958 Epoch 81/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3031 - accuracy: 0.1086 - val_loss: 2.3020 - val_accuracy: 0.1135 Epoch 82/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1079 - val_loss: 2.3029 - val_accuracy: 0.1032 Epoch 83/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3031 - accuracy: 0.1070 - val_loss: 2.3019 - val_accuracy: 0.1135 Epoch 84/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1085 - val_loss: 2.3017 - val_accuracy: 0.1135 Epoch 85/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1099 - val_loss: 2.3034 - val_accuracy: 0.1135 Epoch 86/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1091 - val_loss: 2.3027 - val_accuracy: 0.1135 Epoch 87/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3032 - accuracy: 0.1084 - val_loss: 2.3026 - val_accuracy: 0.1009 Epoch 88/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3033 - accuracy: 0.1070 - val_loss: 2.3029 - val_accuracy: 0.1135 Epoch 89/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1088 - val_loss: 2.3029 - val_accuracy: 0.1028 Epoch 90/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1081 - val_loss: 2.3024 - val_accuracy: 0.1032 Epoch 91/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3031 - accuracy: 0.1082 - val_loss: 2.3038 - val_accuracy: 0.0958 Epoch 92/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1075 - val_loss: 2.3031 - val_accuracy: 0.1028 Epoch 93/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1071 - val_loss: 2.3022 - val_accuracy: 0.1135 Epoch 94/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1081 - val_loss: 2.3023 - val_accuracy: 0.1135 Epoch 95/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1088 - val_loss: 2.3038 - val_accuracy: 0.1135 Epoch 96/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1092 - val_loss: 2.3024 - val_accuracy: 0.1135 Epoch 97/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1086 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 98/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1089 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 99/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3027 - accuracy: 0.1090 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 100/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3027 - accuracy: 0.1086 - val_loss: 2.3033 - val_accuracy: 0.1032 Epoch 101/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1071 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 102/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1083 - val_loss: 2.3024 - val_accuracy: 0.1009 Epoch 103/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3027 - accuracy: 0.1068 - val_loss: 2.3022 - val_accuracy: 0.1135 Epoch 104/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3028 - accuracy: 0.1095 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 105/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3028 - accuracy: 0.1088 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 106/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1104 - val_loss: 2.3017 - val_accuracy: 0.1135 Epoch 107/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1089 - val_loss: 2.3027 - val_accuracy: 0.1135 Epoch 108/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1083 - val_loss: 2.3025 - val_accuracy: 0.1135 Epoch 109/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1098 - val_loss: 2.3041 - val_accuracy: 0.1028 Epoch 110/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1078 - val_loss: 2.3035 - val_accuracy: 0.1135 Epoch 111/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1088 - val_loss: 2.3019 - val_accuracy: 0.1135 Epoch 112/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1100 - val_loss: 2.3035 - val_accuracy: 0.1135 Epoch 113/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1079 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 114/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3027 - accuracy: 0.1089 - val_loss: 2.3027 - val_accuracy: 0.0982 Epoch 115/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1091 - val_loss: 2.3026 - val_accuracy: 0.1009 Epoch 116/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1090 - val_loss: 2.3019 - val_accuracy: 0.1135 Epoch 117/200 938/938 [==============================] - 3s 4ms/step - loss: 2.3025 - accuracy: 0.1092 - val_loss: 2.3033 - val_accuracy: 0.1135 Epoch 118/200 938/938 [==============================] - 3s 4ms/step - loss: 2.3027 - accuracy: 0.1085 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 119/200 938/938 [==============================] - 3s 4ms/step - loss: 2.3026 - accuracy: 0.1101 - val_loss: 2.3022 - val_accuracy: 0.1135 Epoch 120/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1092 - val_loss: 2.3027 - val_accuracy: 0.1135 Epoch 121/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1110 - val_loss: 2.3026 - val_accuracy: 0.1135 Epoch 122/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1095 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 123/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1092 - val_loss: 2.3033 - val_accuracy: 0.0980 Epoch 124/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1107 - val_loss: 2.3019 - val_accuracy: 0.1135 Epoch 125/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1099 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 126/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1093 - val_loss: 2.3024 - val_accuracy: 0.1028 Epoch 127/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1091 - val_loss: 2.3017 - val_accuracy: 0.1135 Epoch 128/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1091 - val_loss: 2.3028 - val_accuracy: 0.1135 Epoch 129/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1092 - val_loss: 2.3021 - val_accuracy: 0.1135 Epoch 130/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1092 - val_loss: 2.3025 - val_accuracy: 0.1135 Epoch 131/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1097 - val_loss: 2.3019 - val_accuracy: 0.1028 Epoch 132/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1078 - val_loss: 2.3021 - val_accuracy: 0.1135 Epoch 133/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1099 - val_loss: 2.3037 - val_accuracy: 0.1135 Epoch 134/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1102 - val_loss: 2.3027 - val_accuracy: 0.1135 Epoch 135/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1095 - val_loss: 2.3020 - val_accuracy: 0.1135 Epoch 136/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1107 - val_loss: 2.3021 - val_accuracy: 0.1135 Epoch 137/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1120 - val_loss: 2.3026 - val_accuracy: 0.1009 Epoch 138/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1090 - val_loss: 2.3024 - val_accuracy: 0.1135 Epoch 139/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1098 - val_loss: 2.3017 - val_accuracy: 0.1135 Epoch 140/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1102 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 141/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1101 - val_loss: 2.3017 - val_accuracy: 0.1010 Epoch 142/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1093 - val_loss: 2.3025 - val_accuracy: 0.1135 Epoch 143/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1089 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 144/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3024 - accuracy: 0.1089 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 145/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1095 - val_loss: 2.3025 - val_accuracy: 0.1010 Epoch 146/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1103 - val_loss: 2.3022 - val_accuracy: 0.1135 Epoch 147/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1102 - val_loss: 2.3022 - val_accuracy: 0.1135 Epoch 148/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1102 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 149/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1105 - val_loss: 2.3017 - val_accuracy: 0.1135 Epoch 150/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1111 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 151/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1093 - val_loss: 2.3023 - val_accuracy: 0.1009 Epoch 152/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1086 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 153/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1106 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 154/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1110 - val_loss: 2.3024 - val_accuracy: 0.1135 Epoch 155/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1107 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 156/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1105 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 157/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1099 - val_loss: 2.3032 - val_accuracy: 0.1028 Epoch 158/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1114 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 159/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1099 - val_loss: 2.3027 - val_accuracy: 0.1028 Epoch 160/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1096 - val_loss: 2.3023 - val_accuracy: 0.1135 Epoch 161/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1101 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 162/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1118 - val_loss: 2.3017 - val_accuracy: 0.1135 Epoch 163/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3023 - accuracy: 0.1110 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 164/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1103 - val_loss: 2.3013 - val_accuracy: 0.1135 Epoch 165/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1097 - val_loss: 2.3027 - val_accuracy: 0.1135 Epoch 166/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1106 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 167/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1111 - val_loss: 2.3017 - val_accuracy: 0.1028 Epoch 168/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1109 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 169/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1108 - val_loss: 2.3011 - val_accuracy: 0.1135 Epoch 170/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1105 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 171/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1103 - val_loss: 2.3023 - val_accuracy: 0.1135 Epoch 172/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1102 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 173/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1097 - val_loss: 2.3017 - val_accuracy: 0.1135 Epoch 174/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1103 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 175/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1117 - val_loss: 2.3023 - val_accuracy: 0.1135 Epoch 176/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1111 - val_loss: 2.3015 - val_accuracy: 0.1028 Epoch 177/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1102 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 178/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1101 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 179/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1115 - val_loss: 2.3012 - val_accuracy: 0.1135 Epoch 180/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1110 - val_loss: 2.3012 - val_accuracy: 0.1135 Epoch 181/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1114 - val_loss: 2.3015 - val_accuracy: 0.1028 Epoch 182/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1101 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 183/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1113 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 184/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1103 - val_loss: 2.3012 - val_accuracy: 0.1135 Epoch 185/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1106 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 186/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1107 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 187/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1114 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 188/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1112 - val_loss: 2.3019 - val_accuracy: 0.1010 Epoch 189/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3021 - accuracy: 0.1101 - val_loss: 2.3014 - val_accuracy: 0.1135 Epoch 190/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1113 - val_loss: 2.3013 - val_accuracy: 0.1135 Epoch 191/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1098 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 192/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3022 - accuracy: 0.1111 - val_loss: 2.3018 - val_accuracy: 0.1135 Epoch 193/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1111 - val_loss: 2.3012 - val_accuracy: 0.1135 Epoch 194/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3018 - accuracy: 0.1113 - val_loss: 2.3022 - val_accuracy: 0.1028 Epoch 195/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3018 - accuracy: 0.1106 - val_loss: 2.3020 - val_accuracy: 0.1028 Epoch 196/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1105 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 197/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1102 - val_loss: 2.3013 - val_accuracy: 0.1135 Epoch 198/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3020 - accuracy: 0.1114 - val_loss: 2.3017 - val_accuracy: 0.1028 Epoch 199/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1107 - val_loss: 2.3016 - val_accuracy: 0.1135 Epoch 200/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3019 - accuracy: 0.1113 - val_loss: 2.3017 - val_accuracy: 0.1135
(b) Activation function: the logistic sigmoid function; initialization: Xavier initializer¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.GlorotNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_9" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_9 (Flatten) (None, 784) 0 _________________________________________________________________ dense_54 (Dense) (None, 512) 401920 _________________________________________________________________ dense_55 (Dense) (None, 512) 262656 _________________________________________________________________ dense_56 (Dense) (None, 512) 262656 _________________________________________________________________ dense_57 (Dense) (None, 512) 262656 _________________________________________________________________ dense_58 (Dense) (None, 512) 262656 _________________________________________________________________ dense_59 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.SGD(learning_rate=.01)
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
history1 = model.fit(X_train, y_train, batch_size=64, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3078 - accuracy: 0.1043 - val_loss: 2.3109 - val_accuracy: 0.1135 Epoch 2/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3065 - accuracy: 0.1063 - val_loss: 2.3083 - val_accuracy: 0.1010 Epoch 3/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3066 - accuracy: 0.1057 - val_loss: 2.3127 - val_accuracy: 0.1010 Epoch 4/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3067 - accuracy: 0.1056 - val_loss: 2.3029 - val_accuracy: 0.0980 Epoch 5/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3066 - accuracy: 0.1043 - val_loss: 2.3063 - val_accuracy: 0.0958 Epoch 6/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3063 - accuracy: 0.1063 - val_loss: 2.3069 - val_accuracy: 0.1028 Epoch 7/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3062 - accuracy: 0.1041 - val_loss: 2.3083 - val_accuracy: 0.1010 Epoch 8/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3061 - accuracy: 0.1036 - val_loss: 2.3062 - val_accuracy: 0.1135 Epoch 9/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3056 - accuracy: 0.1057 - val_loss: 2.3031 - val_accuracy: 0.0982 Epoch 10/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3060 - accuracy: 0.1056 - val_loss: 2.3060 - val_accuracy: 0.1135 Epoch 11/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3056 - accuracy: 0.1049 - val_loss: 2.3063 - val_accuracy: 0.0958 Epoch 12/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3057 - accuracy: 0.1062 - val_loss: 2.3097 - val_accuracy: 0.1009 Epoch 13/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3055 - accuracy: 0.1071 - val_loss: 2.3076 - val_accuracy: 0.1032 Epoch 14/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3054 - accuracy: 0.1047 - val_loss: 2.3024 - val_accuracy: 0.1135 Epoch 15/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3056 - accuracy: 0.1054 - val_loss: 2.3071 - val_accuracy: 0.1028 Epoch 16/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3050 - accuracy: 0.1077 - val_loss: 2.3015 - val_accuracy: 0.1135 Epoch 17/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3054 - accuracy: 0.1076 - val_loss: 2.3100 - val_accuracy: 0.1028 Epoch 18/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3048 - accuracy: 0.1055 - val_loss: 2.3055 - val_accuracy: 0.1862 Epoch 19/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3050 - accuracy: 0.1046 - val_loss: 2.3042 - val_accuracy: 0.1135 Epoch 20/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3045 - accuracy: 0.1088 - val_loss: 2.3024 - val_accuracy: 0.1135 Epoch 21/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3046 - accuracy: 0.1057 - val_loss: 2.3041 - val_accuracy: 0.1032 Epoch 22/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3043 - accuracy: 0.1084 - val_loss: 2.3047 - val_accuracy: 0.1135 Epoch 23/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3040 - accuracy: 0.1079 - val_loss: 2.3143 - val_accuracy: 0.0974 Epoch 24/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3042 - accuracy: 0.1086 - val_loss: 2.3024 - val_accuracy: 0.1135 Epoch 25/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3040 - accuracy: 0.1067 - val_loss: 2.3039 - val_accuracy: 0.1135 Epoch 26/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3038 - accuracy: 0.1090 - val_loss: 2.3009 - val_accuracy: 0.1032 Epoch 27/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3040 - accuracy: 0.1088 - val_loss: 2.3058 - val_accuracy: 0.1135 Epoch 28/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3032 - accuracy: 0.1098 - val_loss: 2.3066 - val_accuracy: 0.1028 Epoch 29/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3032 - accuracy: 0.1109 - val_loss: 2.3050 - val_accuracy: 0.1135 Epoch 30/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3033 - accuracy: 0.1070 - val_loss: 2.3055 - val_accuracy: 0.1028 Epoch 31/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3031 - accuracy: 0.1101 - val_loss: 2.3072 - val_accuracy: 0.1028 Epoch 32/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3029 - accuracy: 0.1086 - val_loss: 2.3042 - val_accuracy: 0.1010 Epoch 33/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3030 - accuracy: 0.1120 - val_loss: 2.3060 - val_accuracy: 0.1028 Epoch 34/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1099 - val_loss: 2.3001 - val_accuracy: 0.1135 Epoch 35/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3026 - accuracy: 0.1111 - val_loss: 2.3060 - val_accuracy: 0.1135 Epoch 36/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3025 - accuracy: 0.1119 - val_loss: 2.2992 - val_accuracy: 0.1135 Epoch 37/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3018 - accuracy: 0.1126 - val_loss: 2.3028 - val_accuracy: 0.2090 Epoch 38/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3015 - accuracy: 0.1121 - val_loss: 2.3025 - val_accuracy: 0.0958 Epoch 39/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3013 - accuracy: 0.1155 - val_loss: 2.3008 - val_accuracy: 0.1135 Epoch 40/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3009 - accuracy: 0.1172 - val_loss: 2.3027 - val_accuracy: 0.1135 Epoch 41/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3004 - accuracy: 0.1160 - val_loss: 2.3000 - val_accuracy: 0.1197 Epoch 42/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2997 - accuracy: 0.1182 - val_loss: 2.3017 - val_accuracy: 0.0958 Epoch 43/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2996 - accuracy: 0.1152 - val_loss: 2.3019 - val_accuracy: 0.0958 Epoch 44/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2995 - accuracy: 0.1187 - val_loss: 2.3006 - val_accuracy: 0.1010 Epoch 45/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2988 - accuracy: 0.1181 - val_loss: 2.2976 - val_accuracy: 0.1135 Epoch 46/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2979 - accuracy: 0.1249 - val_loss: 2.2972 - val_accuracy: 0.1135 Epoch 47/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2971 - accuracy: 0.1240 - val_loss: 2.2986 - val_accuracy: 0.1030 Epoch 48/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2958 - accuracy: 0.1275 - val_loss: 2.2934 - val_accuracy: 0.1135 Epoch 49/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2946 - accuracy: 0.1323 - val_loss: 2.2968 - val_accuracy: 0.1634 Epoch 50/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2930 - accuracy: 0.1366 - val_loss: 2.2911 - val_accuracy: 0.2009 Epoch 51/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2905 - accuracy: 0.1453 - val_loss: 2.2896 - val_accuracy: 0.1032 Epoch 52/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2876 - accuracy: 0.1502 - val_loss: 2.2825 - val_accuracy: 0.1471 Epoch 53/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2826 - accuracy: 0.1672 - val_loss: 2.2802 - val_accuracy: 0.1670 Epoch 54/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2749 - accuracy: 0.1927 - val_loss: 2.2700 - val_accuracy: 0.1900 Epoch 55/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2605 - accuracy: 0.2166 - val_loss: 2.2563 - val_accuracy: 0.1463 Epoch 56/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2296 - accuracy: 0.2388 - val_loss: 2.2019 - val_accuracy: 0.2626 Epoch 57/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1527 - accuracy: 0.2533 - val_loss: 2.0813 - val_accuracy: 0.2496 Epoch 58/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9955 - accuracy: 0.2742 - val_loss: 1.9011 - val_accuracy: 0.2962 Epoch 59/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8542 - accuracy: 0.3022 - val_loss: 1.7988 - val_accuracy: 0.3156 Epoch 60/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7794 - accuracy: 0.3248 - val_loss: 1.7413 - val_accuracy: 0.3226 Epoch 61/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7257 - accuracy: 0.3378 - val_loss: 1.6775 - val_accuracy: 0.3694 Epoch 62/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6731 - accuracy: 0.3541 - val_loss: 1.6257 - val_accuracy: 0.3679 Epoch 63/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6298 - accuracy: 0.3687 - val_loss: 1.5857 - val_accuracy: 0.3675 Epoch 64/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6003 - accuracy: 0.3732 - val_loss: 1.5648 - val_accuracy: 0.3718 Epoch 65/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5805 - accuracy: 0.3830 - val_loss: 1.5520 - val_accuracy: 0.3790 Epoch 66/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5661 - accuracy: 0.3851 - val_loss: 1.5336 - val_accuracy: 0.3942 Epoch 67/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5546 - accuracy: 0.3906 - val_loss: 1.5315 - val_accuracy: 0.3868 Epoch 68/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5430 - accuracy: 0.3971 - val_loss: 1.5151 - val_accuracy: 0.4122 Epoch 69/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5323 - accuracy: 0.4053 - val_loss: 1.5072 - val_accuracy: 0.3905 Epoch 70/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5216 - accuracy: 0.4084 - val_loss: 1.4968 - val_accuracy: 0.4075 Epoch 71/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5088 - accuracy: 0.4168 - val_loss: 1.4868 - val_accuracy: 0.4135 Epoch 72/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4956 - accuracy: 0.4264 - val_loss: 1.4746 - val_accuracy: 0.4447 Epoch 73/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4807 - accuracy: 0.4370 - val_loss: 1.4494 - val_accuracy: 0.4352 Epoch 74/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4633 - accuracy: 0.4509 - val_loss: 1.4287 - val_accuracy: 0.4682 Epoch 75/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4409 - accuracy: 0.4690 - val_loss: 1.4030 - val_accuracy: 0.4987 Epoch 76/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4085 - accuracy: 0.4946 - val_loss: 1.3632 - val_accuracy: 0.5318 Epoch 77/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3549 - accuracy: 0.5363 - val_loss: 1.2959 - val_accuracy: 0.5382 Epoch 78/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2619 - accuracy: 0.5797 - val_loss: 1.1798 - val_accuracy: 0.6005 Epoch 79/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1458 - accuracy: 0.6191 - val_loss: 1.0769 - val_accuracy: 0.6412 Epoch 80/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0613 - accuracy: 0.6472 - val_loss: 1.0075 - val_accuracy: 0.6533 Epoch 81/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0103 - accuracy: 0.6638 - val_loss: 0.9783 - val_accuracy: 0.6616 Epoch 82/200 938/938 [==============================] - 3s 3ms/step - loss: 0.9736 - accuracy: 0.6785 - val_loss: 0.9476 - val_accuracy: 0.6707 Epoch 83/200 938/938 [==============================] - 3s 3ms/step - loss: 0.9405 - accuracy: 0.6907 - val_loss: 0.9224 - val_accuracy: 0.6842 Epoch 84/200 938/938 [==============================] - 3s 3ms/step - loss: 0.9068 - accuracy: 0.7056 - val_loss: 0.8669 - val_accuracy: 0.7058 Epoch 85/200 938/938 [==============================] - 3s 3ms/step - loss: 0.8686 - accuracy: 0.7210 - val_loss: 0.8313 - val_accuracy: 0.7229 Epoch 86/200 938/938 [==============================] - 3s 3ms/step - loss: 0.8234 - accuracy: 0.7376 - val_loss: 0.7844 - val_accuracy: 0.7443 Epoch 87/200 938/938 [==============================] - 3s 3ms/step - loss: 0.7732 - accuracy: 0.7566 - val_loss: 0.7335 - val_accuracy: 0.7673 Epoch 88/200 938/938 [==============================] - 3s 3ms/step - loss: 0.7268 - accuracy: 0.7723 - val_loss: 0.7055 - val_accuracy: 0.7721 Epoch 89/200 938/938 [==============================] - 3s 3ms/step - loss: 0.6904 - accuracy: 0.7830 - val_loss: 0.6622 - val_accuracy: 0.7926 Epoch 90/200 938/938 [==============================] - 3s 3ms/step - loss: 0.6636 - accuracy: 0.7938 - val_loss: 0.6338 - val_accuracy: 0.8043 Epoch 91/200 938/938 [==============================] - 3s 3ms/step - loss: 0.6422 - accuracy: 0.8009 - val_loss: 0.6340 - val_accuracy: 0.7997 Epoch 92/200 938/938 [==============================] - 3s 3ms/step - loss: 0.6244 - accuracy: 0.8076 - val_loss: 0.6242 - val_accuracy: 0.7975 Epoch 93/200 938/938 [==============================] - 3s 3ms/step - loss: 0.6093 - accuracy: 0.8151 - val_loss: 0.5909 - val_accuracy: 0.8207 Epoch 94/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5950 - accuracy: 0.8202 - val_loss: 0.5760 - val_accuracy: 0.8255 Epoch 95/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5828 - accuracy: 0.8255 - val_loss: 0.5644 - val_accuracy: 0.8318 Epoch 96/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5703 - accuracy: 0.8304 - val_loss: 0.5515 - val_accuracy: 0.8367 Epoch 97/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5592 - accuracy: 0.8349 - val_loss: 0.5392 - val_accuracy: 0.8405 Epoch 98/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5491 - accuracy: 0.8394 - val_loss: 0.5353 - val_accuracy: 0.8445 Epoch 99/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5397 - accuracy: 0.8419 - val_loss: 0.5231 - val_accuracy: 0.8466 Epoch 100/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5297 - accuracy: 0.8458 - val_loss: 0.5143 - val_accuracy: 0.8530 Epoch 101/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5214 - accuracy: 0.8497 - val_loss: 0.5245 - val_accuracy: 0.8493 Epoch 102/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5136 - accuracy: 0.8531 - val_loss: 0.5037 - val_accuracy: 0.8539 Epoch 103/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5059 - accuracy: 0.8557 - val_loss: 0.4947 - val_accuracy: 0.8571 Epoch 104/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4977 - accuracy: 0.8586 - val_loss: 0.4852 - val_accuracy: 0.8637 Epoch 105/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4913 - accuracy: 0.8613 - val_loss: 0.4754 - val_accuracy: 0.8682 Epoch 106/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4850 - accuracy: 0.8640 - val_loss: 0.4748 - val_accuracy: 0.8704 Epoch 107/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4781 - accuracy: 0.8665 - val_loss: 0.4700 - val_accuracy: 0.8705 Epoch 108/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4719 - accuracy: 0.8679 - val_loss: 0.4809 - val_accuracy: 0.8647 Epoch 109/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4657 - accuracy: 0.8705 - val_loss: 0.4564 - val_accuracy: 0.8752 Epoch 110/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4600 - accuracy: 0.8723 - val_loss: 0.4489 - val_accuracy: 0.8786 Epoch 111/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4549 - accuracy: 0.8738 - val_loss: 0.4701 - val_accuracy: 0.8702 Epoch 112/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4486 - accuracy: 0.8762 - val_loss: 0.4378 - val_accuracy: 0.8817 Epoch 113/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4428 - accuracy: 0.8782 - val_loss: 0.4334 - val_accuracy: 0.8829 Epoch 114/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4370 - accuracy: 0.8805 - val_loss: 0.4373 - val_accuracy: 0.8796 Epoch 115/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4326 - accuracy: 0.8811 - val_loss: 0.4236 - val_accuracy: 0.8865 Epoch 116/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4274 - accuracy: 0.8837 - val_loss: 0.4328 - val_accuracy: 0.8840 Epoch 117/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4224 - accuracy: 0.8851 - val_loss: 0.4269 - val_accuracy: 0.8833 Epoch 118/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4174 - accuracy: 0.8873 - val_loss: 0.4113 - val_accuracy: 0.8915 Epoch 119/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4122 - accuracy: 0.8884 - val_loss: 0.4101 - val_accuracy: 0.8896 Epoch 120/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4077 - accuracy: 0.8899 - val_loss: 0.4069 - val_accuracy: 0.8916 Epoch 121/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4035 - accuracy: 0.8911 - val_loss: 0.4194 - val_accuracy: 0.8853 Epoch 122/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3986 - accuracy: 0.8924 - val_loss: 0.3988 - val_accuracy: 0.8939 Epoch 123/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3946 - accuracy: 0.8935 - val_loss: 0.4163 - val_accuracy: 0.8858 Epoch 124/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3895 - accuracy: 0.8949 - val_loss: 0.3929 - val_accuracy: 0.8972 Epoch 125/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3846 - accuracy: 0.8968 - val_loss: 0.3888 - val_accuracy: 0.8982 Epoch 126/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3808 - accuracy: 0.8977 - val_loss: 0.4029 - val_accuracy: 0.8915 Epoch 127/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3759 - accuracy: 0.8989 - val_loss: 0.3966 - val_accuracy: 0.8919 Epoch 128/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3723 - accuracy: 0.8994 - val_loss: 0.3925 - val_accuracy: 0.8953 Epoch 129/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3679 - accuracy: 0.9016 - val_loss: 0.3693 - val_accuracy: 0.9032 Epoch 130/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3632 - accuracy: 0.9031 - val_loss: 0.3832 - val_accuracy: 0.8978 Epoch 131/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3598 - accuracy: 0.9041 - val_loss: 0.3869 - val_accuracy: 0.8965 Epoch 132/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3557 - accuracy: 0.9049 - val_loss: 0.3652 - val_accuracy: 0.9063 Epoch 133/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3509 - accuracy: 0.9069 - val_loss: 0.3900 - val_accuracy: 0.8952 Epoch 134/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3479 - accuracy: 0.9076 - val_loss: 0.3610 - val_accuracy: 0.9069 Epoch 135/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3433 - accuracy: 0.9093 - val_loss: 0.3489 - val_accuracy: 0.9107 Epoch 136/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3400 - accuracy: 0.9094 - val_loss: 0.3513 - val_accuracy: 0.9097 Epoch 137/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3356 - accuracy: 0.9115 - val_loss: 0.3502 - val_accuracy: 0.9098 Epoch 138/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3326 - accuracy: 0.9129 - val_loss: 0.3407 - val_accuracy: 0.9136 Epoch 139/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3288 - accuracy: 0.9134 - val_loss: 0.3536 - val_accuracy: 0.9069 Epoch 140/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3232 - accuracy: 0.9154 - val_loss: 0.3444 - val_accuracy: 0.9112 Epoch 141/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3212 - accuracy: 0.9161 - val_loss: 0.3433 - val_accuracy: 0.9108 Epoch 142/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3179 - accuracy: 0.9166 - val_loss: 0.3308 - val_accuracy: 0.9177 Epoch 143/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3137 - accuracy: 0.9175 - val_loss: 0.3549 - val_accuracy: 0.9066 Epoch 144/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3104 - accuracy: 0.9185 - val_loss: 0.3260 - val_accuracy: 0.9178 Epoch 145/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3057 - accuracy: 0.9198 - val_loss: 0.3258 - val_accuracy: 0.9156 Epoch 146/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3030 - accuracy: 0.9194 - val_loss: 0.3197 - val_accuracy: 0.9185 Epoch 147/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2989 - accuracy: 0.9215 - val_loss: 0.3427 - val_accuracy: 0.9110 Epoch 148/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2956 - accuracy: 0.9221 - val_loss: 0.3324 - val_accuracy: 0.9125 Epoch 149/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2921 - accuracy: 0.9239 - val_loss: 0.3194 - val_accuracy: 0.9185 Epoch 150/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2884 - accuracy: 0.9237 - val_loss: 0.3103 - val_accuracy: 0.9199 Epoch 151/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2849 - accuracy: 0.9253 - val_loss: 0.3444 - val_accuracy: 0.9085 Epoch 152/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2813 - accuracy: 0.9261 - val_loss: 0.3023 - val_accuracy: 0.9224 Epoch 153/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2791 - accuracy: 0.9266 - val_loss: 0.3059 - val_accuracy: 0.9220 Epoch 154/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2754 - accuracy: 0.9270 - val_loss: 0.3060 - val_accuracy: 0.9196 Epoch 155/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2702 - accuracy: 0.9283 - val_loss: 0.2915 - val_accuracy: 0.9247 Epoch 156/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2682 - accuracy: 0.9288 - val_loss: 0.2869 - val_accuracy: 0.9270 Epoch 157/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2636 - accuracy: 0.9307 - val_loss: 0.2838 - val_accuracy: 0.9271 Epoch 158/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2601 - accuracy: 0.9314 - val_loss: 0.2892 - val_accuracy: 0.9259 Epoch 159/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2575 - accuracy: 0.9320 - val_loss: 0.2869 - val_accuracy: 0.9264 Epoch 160/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2558 - accuracy: 0.9329 - val_loss: 0.2837 - val_accuracy: 0.9280 Epoch 161/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2505 - accuracy: 0.9336 - val_loss: 0.2856 - val_accuracy: 0.9248 Epoch 162/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2479 - accuracy: 0.9340 - val_loss: 0.2721 - val_accuracy: 0.9287 Epoch 163/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2451 - accuracy: 0.9350 - val_loss: 0.2701 - val_accuracy: 0.9303 Epoch 164/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2420 - accuracy: 0.9364 - val_loss: 0.2861 - val_accuracy: 0.9244 Epoch 165/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2388 - accuracy: 0.9362 - val_loss: 0.3065 - val_accuracy: 0.9186 Epoch 166/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2356 - accuracy: 0.9377 - val_loss: 0.2721 - val_accuracy: 0.9285 Epoch 167/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2323 - accuracy: 0.9389 - val_loss: 0.2758 - val_accuracy: 0.9277 Epoch 168/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2285 - accuracy: 0.9392 - val_loss: 0.2881 - val_accuracy: 0.9227 Epoch 169/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2257 - accuracy: 0.9406 - val_loss: 0.2537 - val_accuracy: 0.9350 Epoch 170/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2238 - accuracy: 0.9406 - val_loss: 0.2580 - val_accuracy: 0.9321 Epoch 171/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2198 - accuracy: 0.9415 - val_loss: 0.2459 - val_accuracy: 0.9356 Epoch 172/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2172 - accuracy: 0.9417 - val_loss: 0.2496 - val_accuracy: 0.9356 Epoch 173/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2142 - accuracy: 0.9432 - val_loss: 0.2442 - val_accuracy: 0.9357 Epoch 174/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2123 - accuracy: 0.9436 - val_loss: 0.2440 - val_accuracy: 0.9355 Epoch 175/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2082 - accuracy: 0.9446 - val_loss: 0.2395 - val_accuracy: 0.9357 Epoch 176/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2059 - accuracy: 0.9456 - val_loss: 0.2414 - val_accuracy: 0.9361 Epoch 177/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2038 - accuracy: 0.9456 - val_loss: 0.2298 - val_accuracy: 0.9388 Epoch 178/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2008 - accuracy: 0.9465 - val_loss: 0.2290 - val_accuracy: 0.9396 Epoch 179/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1983 - accuracy: 0.9470 - val_loss: 0.2399 - val_accuracy: 0.9366 Epoch 180/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1956 - accuracy: 0.9477 - val_loss: 0.2264 - val_accuracy: 0.9401 Epoch 181/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1917 - accuracy: 0.9485 - val_loss: 0.2225 - val_accuracy: 0.9406 Epoch 182/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1902 - accuracy: 0.9494 - val_loss: 0.2237 - val_accuracy: 0.9400 Epoch 183/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1868 - accuracy: 0.9506 - val_loss: 0.2181 - val_accuracy: 0.9420 Epoch 184/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1843 - accuracy: 0.9506 - val_loss: 0.2160 - val_accuracy: 0.9422 Epoch 185/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1821 - accuracy: 0.9512 - val_loss: 0.2597 - val_accuracy: 0.9315 Epoch 186/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1795 - accuracy: 0.9521 - val_loss: 0.2176 - val_accuracy: 0.9431 Epoch 187/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1781 - accuracy: 0.9521 - val_loss: 0.2089 - val_accuracy: 0.9449 Epoch 188/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1750 - accuracy: 0.9529 - val_loss: 0.2172 - val_accuracy: 0.9392 Epoch 189/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1733 - accuracy: 0.9531 - val_loss: 0.2328 - val_accuracy: 0.9391 Epoch 190/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1713 - accuracy: 0.9540 - val_loss: 0.2079 - val_accuracy: 0.9446 Epoch 191/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1690 - accuracy: 0.9548 - val_loss: 0.2239 - val_accuracy: 0.9417 Epoch 192/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1680 - accuracy: 0.9545 - val_loss: 0.2346 - val_accuracy: 0.9357 Epoch 193/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1647 - accuracy: 0.9547 - val_loss: 0.2040 - val_accuracy: 0.9447 Epoch 194/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1621 - accuracy: 0.9561 - val_loss: 0.2155 - val_accuracy: 0.9436 Epoch 195/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1604 - accuracy: 0.9564 - val_loss: 0.1947 - val_accuracy: 0.9480 Epoch 196/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1579 - accuracy: 0.9571 - val_loss: 0.1982 - val_accuracy: 0.9480 Epoch 197/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1565 - accuracy: 0.9575 - val_loss: 0.2075 - val_accuracy: 0.9449 Epoch 198/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1543 - accuracy: 0.9581 - val_loss: 0.1888 - val_accuracy: 0.9484 Epoch 199/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1515 - accuracy: 0.9585 - val_loss: 0.1892 - val_accuracy: 0.9480 Epoch 200/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1503 - accuracy: 0.9587 - val_loss: 0.1925 - val_accuracy: 0.9485
(c) Activation function: ReLU; initialization: random numbers generated from the normal¶
distribution (μ = 0, σ = 0.01)
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.01, seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_10" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_10 (Flatten) (None, 784) 0 _________________________________________________________________ dense_60 (Dense) (None, 512) 401920 _________________________________________________________________ dense_61 (Dense) (None, 512) 262656 _________________________________________________________________ dense_62 (Dense) (None, 512) 262656 _________________________________________________________________ dense_63 (Dense) (None, 512) 262656 _________________________________________________________________ dense_64 (Dense) (None, 512) 262656 _________________________________________________________________ dense_65 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.SGD(learning_rate=.01)
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
history2 = model.fit(X_train, y_train, batch_size=64, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3018 - accuracy: 0.1120 - val_loss: 2.3013 - val_accuracy: 0.1135 Epoch 2/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3013 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 3/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 4/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 5/200 938/938 [==============================] - 3s 4ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 6/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 7/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 8/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 9/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 10/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 11/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 12/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 13/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 14/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 15/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 16/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 17/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 18/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 19/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 20/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 21/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 22/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 23/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 24/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 25/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 26/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 27/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3012 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 28/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 29/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 30/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3010 - val_accuracy: 0.1135 Epoch 31/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 32/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 33/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 34/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 35/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 36/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 37/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 38/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3009 - val_accuracy: 0.1135 Epoch 39/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3008 - val_accuracy: 0.1135 Epoch 40/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3011 - accuracy: 0.1124 - val_loss: 2.3008 - val_accuracy: 0.1135 Epoch 41/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3010 - accuracy: 0.1124 - val_loss: 2.3008 - val_accuracy: 0.1135 Epoch 42/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3010 - accuracy: 0.1124 - val_loss: 2.3008 - val_accuracy: 0.1135 Epoch 43/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3010 - accuracy: 0.1124 - val_loss: 2.3008 - val_accuracy: 0.1135 Epoch 44/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3010 - accuracy: 0.1124 - val_loss: 2.3008 - val_accuracy: 0.1135 Epoch 45/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3010 - accuracy: 0.1124 - val_loss: 2.3007 - val_accuracy: 0.1135 Epoch 46/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3009 - accuracy: 0.1124 - val_loss: 2.3007 - val_accuracy: 0.1135 Epoch 47/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3009 - accuracy: 0.1124 - val_loss: 2.3007 - val_accuracy: 0.1135 Epoch 48/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3008 - accuracy: 0.1124 - val_loss: 2.3006 - val_accuracy: 0.1135 Epoch 49/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3008 - accuracy: 0.1124 - val_loss: 2.3005 - val_accuracy: 0.1135 Epoch 50/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3007 - accuracy: 0.1124 - val_loss: 2.3004 - val_accuracy: 0.1135 Epoch 51/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3006 - accuracy: 0.1124 - val_loss: 2.3003 - val_accuracy: 0.1135 Epoch 52/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3004 - accuracy: 0.1124 - val_loss: 2.3001 - val_accuracy: 0.1135 Epoch 53/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3001 - accuracy: 0.1124 - val_loss: 2.2997 - val_accuracy: 0.1135 Epoch 54/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2997 - accuracy: 0.1124 - val_loss: 2.2991 - val_accuracy: 0.1135 Epoch 55/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2987 - accuracy: 0.1124 - val_loss: 2.2975 - val_accuracy: 0.1135 Epoch 56/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2959 - accuracy: 0.1124 - val_loss: 2.2922 - val_accuracy: 0.1135 Epoch 57/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2766 - accuracy: 0.1768 - val_loss: 2.2261 - val_accuracy: 0.2105 Epoch 58/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0037 - accuracy: 0.2184 - val_loss: 1.8252 - val_accuracy: 0.2414 Epoch 59/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7324 - accuracy: 0.2667 - val_loss: 1.6546 - val_accuracy: 0.3045 Epoch 60/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5745 - accuracy: 0.3407 - val_loss: 1.4671 - val_accuracy: 0.4449 Epoch 61/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3193 - accuracy: 0.4944 - val_loss: 1.1430 - val_accuracy: 0.5548 Epoch 62/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0701 - accuracy: 0.5969 - val_loss: 1.0867 - val_accuracy: 0.5596 Epoch 63/200 938/938 [==============================] - 3s 3ms/step - loss: 0.9064 - accuracy: 0.6730 - val_loss: 1.0725 - val_accuracy: 0.5881 Epoch 64/200 938/938 [==============================] - 3s 3ms/step - loss: 0.7820 - accuracy: 0.7400 - val_loss: 0.9289 - val_accuracy: 0.6774 Epoch 65/200 938/938 [==============================] - 3s 3ms/step - loss: 0.6609 - accuracy: 0.8132 - val_loss: 0.6117 - val_accuracy: 0.8500 Epoch 66/200 938/938 [==============================] - 3s 3ms/step - loss: 0.5523 - accuracy: 0.8520 - val_loss: 0.5201 - val_accuracy: 0.8765 Epoch 67/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4572 - accuracy: 0.8789 - val_loss: 0.6081 - val_accuracy: 0.8021 Epoch 68/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3870 - accuracy: 0.8978 - val_loss: 0.4010 - val_accuracy: 0.9048 Epoch 69/200 938/938 [==============================] - 3s 3ms/step - loss: 0.3313 - accuracy: 0.9132 - val_loss: 0.4584 - val_accuracy: 0.8751 Epoch 70/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2853 - accuracy: 0.9247 - val_loss: 0.3507 - val_accuracy: 0.9173 Epoch 71/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2494 - accuracy: 0.9324 - val_loss: 0.3447 - val_accuracy: 0.9184 Epoch 72/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2188 - accuracy: 0.9404 - val_loss: 0.4062 - val_accuracy: 0.8928 Epoch 73/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1907 - accuracy: 0.9481 - val_loss: 0.3564 - val_accuracy: 0.9190 Epoch 74/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1693 - accuracy: 0.9533 - val_loss: 0.3268 - val_accuracy: 0.9289 Epoch 75/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1478 - accuracy: 0.9589 - val_loss: 0.3117 - val_accuracy: 0.9292 Epoch 76/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1310 - accuracy: 0.9632 - val_loss: 0.3025 - val_accuracy: 0.9323 Epoch 77/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1156 - accuracy: 0.9685 - val_loss: 0.3100 - val_accuracy: 0.9362 Epoch 78/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1004 - accuracy: 0.9727 - val_loss: 0.3141 - val_accuracy: 0.9379 Epoch 79/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0888 - accuracy: 0.9756 - val_loss: 0.3371 - val_accuracy: 0.9249 Epoch 80/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0793 - accuracy: 0.9786 - val_loss: 0.3176 - val_accuracy: 0.9401 Epoch 81/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0709 - accuracy: 0.9806 - val_loss: 0.3252 - val_accuracy: 0.9358 Epoch 82/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0677 - accuracy: 0.9811 - val_loss: 0.3060 - val_accuracy: 0.9412 Epoch 83/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0575 - accuracy: 0.9839 - val_loss: 0.3431 - val_accuracy: 0.9379 Epoch 84/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0504 - accuracy: 0.9859 - val_loss: 0.3831 - val_accuracy: 0.9302 Epoch 85/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0473 - accuracy: 0.9870 - val_loss: 0.3190 - val_accuracy: 0.9455 Epoch 86/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0445 - accuracy: 0.9875 - val_loss: 0.3115 - val_accuracy: 0.9443 Epoch 87/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0339 - accuracy: 0.9909 - val_loss: 0.3321 - val_accuracy: 0.9461 Epoch 88/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0296 - accuracy: 0.9923 - val_loss: 0.3446 - val_accuracy: 0.9438 Epoch 89/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0278 - accuracy: 0.9926 - val_loss: 0.3556 - val_accuracy: 0.9460 Epoch 90/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0280 - accuracy: 0.9922 - val_loss: 0.4406 - val_accuracy: 0.9300 Epoch 91/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0254 - accuracy: 0.9931 - val_loss: 0.3440 - val_accuracy: 0.9483 Epoch 92/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0198 - accuracy: 0.9948 - val_loss: 0.3568 - val_accuracy: 0.9452 Epoch 93/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0140 - accuracy: 0.9970 - val_loss: 0.3722 - val_accuracy: 0.9478 Epoch 94/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0123 - accuracy: 0.9974 - val_loss: 0.5255 - val_accuracy: 0.9292 Epoch 95/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0107 - accuracy: 0.9979 - val_loss: 0.3867 - val_accuracy: 0.9455 Epoch 96/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0090 - accuracy: 0.9981 - val_loss: 0.3762 - val_accuracy: 0.9485 Epoch 97/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0088 - accuracy: 0.9982 - val_loss: 0.3843 - val_accuracy: 0.9496 Epoch 98/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0078 - accuracy: 0.9986 - val_loss: 0.3736 - val_accuracy: 0.9503 Epoch 99/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0060 - accuracy: 0.9990 - val_loss: 0.3947 - val_accuracy: 0.9496 Epoch 100/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0059 - accuracy: 0.9988 - val_loss: 0.5080 - val_accuracy: 0.9187 Epoch 101/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0056 - accuracy: 0.9991 - val_loss: 0.4122 - val_accuracy: 0.9478 Epoch 102/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0046 - accuracy: 0.9994 - val_loss: 0.4045 - val_accuracy: 0.9508 Epoch 103/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0042 - accuracy: 0.9994 - val_loss: 0.4263 - val_accuracy: 0.9484 Epoch 104/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0058 - accuracy: 0.9988 - val_loss: 0.4234 - val_accuracy: 0.9496 Epoch 105/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0031 - accuracy: 0.9997 - val_loss: 0.4072 - val_accuracy: 0.9511 Epoch 106/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0030 - accuracy: 0.9997 - val_loss: 0.4200 - val_accuracy: 0.9509 Epoch 107/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0031 - accuracy: 0.9995 - val_loss: 0.4367 - val_accuracy: 0.9518 Epoch 108/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0036 - accuracy: 0.9993 - val_loss: 0.4335 - val_accuracy: 0.9503 Epoch 109/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0026 - accuracy: 0.9996 - val_loss: 0.4289 - val_accuracy: 0.9496 Epoch 110/200 938/938 [==============================] - 3s 4ms/step - loss: 0.0088 - accuracy: 0.9980 - val_loss: 0.5239 - val_accuracy: 0.9338 Epoch 111/200 938/938 [==============================] - 3s 4ms/step - loss: 0.0294 - accuracy: 0.9905 - val_loss: 0.4105 - val_accuracy: 0.9459 Epoch 112/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0308 - accuracy: 0.9903 - val_loss: 0.4146 - val_accuracy: 0.9498 Epoch 113/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0235 - accuracy: 0.9923 - val_loss: 0.3854 - val_accuracy: 0.9507 Epoch 114/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0152 - accuracy: 0.9956 - val_loss: 0.5354 - val_accuracy: 0.9352 Epoch 115/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0096 - accuracy: 0.9974 - val_loss: 0.3948 - val_accuracy: 0.9517 Epoch 116/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0044 - accuracy: 0.9992 - val_loss: 0.3994 - val_accuracy: 0.9534 Epoch 117/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0021 - accuracy: 0.9998 - val_loss: 0.4176 - val_accuracy: 0.9529 Epoch 118/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0012 - accuracy: 0.9999 - val_loss: 0.4091 - val_accuracy: 0.9519 Epoch 119/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0012 - accuracy: 0.9999 - val_loss: 0.4275 - val_accuracy: 0.9522 Epoch 120/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0010 - accuracy: 0.9999 - val_loss: 0.4227 - val_accuracy: 0.9533 Epoch 121/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0012 - accuracy: 0.9999 - val_loss: 0.4319 - val_accuracy: 0.9532 Epoch 122/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0011 - accuracy: 0.9999 - val_loss: 0.4293 - val_accuracy: 0.9524 Epoch 123/200 938/938 [==============================] - 3s 3ms/step - loss: 7.2665e-04 - accuracy: 0.9999 - val_loss: 0.4409 - val_accuracy: 0.9524 Epoch 124/200 938/938 [==============================] - 3s 3ms/step - loss: 6.3522e-04 - accuracy: 1.0000 - val_loss: 0.4413 - val_accuracy: 0.9526 Epoch 125/200 938/938 [==============================] - 3s 3ms/step - loss: 5.0149e-04 - accuracy: 1.0000 - val_loss: 0.4458 - val_accuracy: 0.9527 Epoch 126/200 938/938 [==============================] - 3s 3ms/step - loss: 4.7013e-04 - accuracy: 1.0000 - val_loss: 0.4517 - val_accuracy: 0.9526 Epoch 127/200 938/938 [==============================] - 3s 3ms/step - loss: 4.3521e-04 - accuracy: 1.0000 - val_loss: 0.4608 - val_accuracy: 0.9521 Epoch 128/200 938/938 [==============================] - 3s 3ms/step - loss: 4.3093e-04 - accuracy: 1.0000 - val_loss: 0.4525 - val_accuracy: 0.9529 Epoch 129/200 938/938 [==============================] - 3s 3ms/step - loss: 4.0651e-04 - accuracy: 1.0000 - val_loss: 0.4577 - val_accuracy: 0.9523 Epoch 130/200 938/938 [==============================] - 3s 3ms/step - loss: 3.9236e-04 - accuracy: 1.0000 - val_loss: 0.4604 - val_accuracy: 0.9525 Epoch 131/200 938/938 [==============================] - 3s 3ms/step - loss: 3.7346e-04 - accuracy: 1.0000 - val_loss: 0.4635 - val_accuracy: 0.9522 Epoch 132/200 938/938 [==============================] - 3s 3ms/step - loss: 3.6180e-04 - accuracy: 1.0000 - val_loss: 0.4630 - val_accuracy: 0.9529 Epoch 133/200 938/938 [==============================] - 3s 3ms/step - loss: 3.5008e-04 - accuracy: 1.0000 - val_loss: 0.4634 - val_accuracy: 0.9521 Epoch 134/200 938/938 [==============================] - 3s 3ms/step - loss: 3.3836e-04 - accuracy: 1.0000 - val_loss: 0.4753 - val_accuracy: 0.9519 Epoch 135/200 938/938 [==============================] - 3s 3ms/step - loss: 3.3323e-04 - accuracy: 1.0000 - val_loss: 0.4684 - val_accuracy: 0.9521 Epoch 136/200 938/938 [==============================] - 3s 3ms/step - loss: 3.2477e-04 - accuracy: 1.0000 - val_loss: 0.4679 - val_accuracy: 0.9523 Epoch 137/200 938/938 [==============================] - 3s 3ms/step - loss: 3.1256e-04 - accuracy: 1.0000 - val_loss: 0.4728 - val_accuracy: 0.9518 Epoch 138/200 938/938 [==============================] - 3s 3ms/step - loss: 2.9770e-04 - accuracy: 1.0000 - val_loss: 0.4733 - val_accuracy: 0.9524 Epoch 139/200 938/938 [==============================] - 3s 3ms/step - loss: 2.9488e-04 - accuracy: 1.0000 - val_loss: 0.4765 - val_accuracy: 0.9520 Epoch 140/200 938/938 [==============================] - 3s 3ms/step - loss: 2.8132e-04 - accuracy: 1.0000 - val_loss: 0.4774 - val_accuracy: 0.9521 Epoch 141/200 938/938 [==============================] - 3s 3ms/step - loss: 2.7649e-04 - accuracy: 1.0000 - val_loss: 0.4782 - val_accuracy: 0.9521 Epoch 142/200 938/938 [==============================] - 3s 3ms/step - loss: 2.6361e-04 - accuracy: 1.0000 - val_loss: 0.4845 - val_accuracy: 0.9514 Epoch 143/200 938/938 [==============================] - 3s 3ms/step - loss: 2.6136e-04 - accuracy: 1.0000 - val_loss: 0.4796 - val_accuracy: 0.9518 Epoch 144/200 938/938 [==============================] - 3s 3ms/step - loss: 2.4710e-04 - accuracy: 1.0000 - val_loss: 0.4831 - val_accuracy: 0.9518 Epoch 145/200 938/938 [==============================] - 3s 3ms/step - loss: 2.4919e-04 - accuracy: 1.0000 - val_loss: 0.4831 - val_accuracy: 0.9516 Epoch 146/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3890e-04 - accuracy: 1.0000 - val_loss: 0.4849 - val_accuracy: 0.9520 Epoch 147/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2987e-04 - accuracy: 1.0000 - val_loss: 0.4851 - val_accuracy: 0.9517 Epoch 148/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2130e-04 - accuracy: 1.0000 - val_loss: 0.4855 - val_accuracy: 0.9515 Epoch 149/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1278e-04 - accuracy: 1.0000 - val_loss: 0.4867 - val_accuracy: 0.9514 Epoch 150/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8723e-04 - accuracy: 1.0000 - val_loss: 0.4905 - val_accuracy: 0.9510 Epoch 151/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8264e-04 - accuracy: 1.0000 - val_loss: 0.4981 - val_accuracy: 0.9510 Epoch 152/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7668e-04 - accuracy: 1.0000 - val_loss: 0.4923 - val_accuracy: 0.9510 Epoch 153/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6909e-04 - accuracy: 1.0000 - val_loss: 0.4951 - val_accuracy: 0.9514 Epoch 154/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6577e-04 - accuracy: 1.0000 - val_loss: 0.4965 - val_accuracy: 0.9510 Epoch 155/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6164e-04 - accuracy: 1.0000 - val_loss: 0.4991 - val_accuracy: 0.9512 Epoch 156/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5775e-04 - accuracy: 1.0000 - val_loss: 0.4985 - val_accuracy: 0.9513 Epoch 157/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5414e-04 - accuracy: 1.0000 - val_loss: 0.4993 - val_accuracy: 0.9509 Epoch 158/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5093e-04 - accuracy: 1.0000 - val_loss: 0.5011 - val_accuracy: 0.9512 Epoch 159/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4773e-04 - accuracy: 1.0000 - val_loss: 0.5032 - val_accuracy: 0.9512 Epoch 160/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4504e-04 - accuracy: 1.0000 - val_loss: 0.5026 - val_accuracy: 0.9512 Epoch 161/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4218e-04 - accuracy: 1.0000 - val_loss: 0.5032 - val_accuracy: 0.9512 Epoch 162/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3955e-04 - accuracy: 1.0000 - val_loss: 0.5051 - val_accuracy: 0.9513 Epoch 163/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3701e-04 - accuracy: 1.0000 - val_loss: 0.5064 - val_accuracy: 0.9509 Epoch 164/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3440e-04 - accuracy: 1.0000 - val_loss: 0.5055 - val_accuracy: 0.9511 Epoch 165/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3227e-04 - accuracy: 1.0000 - val_loss: 0.5074 - val_accuracy: 0.9511 Epoch 166/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3002e-04 - accuracy: 1.0000 - val_loss: 0.5083 - val_accuracy: 0.9511 Epoch 167/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2773e-04 - accuracy: 1.0000 - val_loss: 0.5096 - val_accuracy: 0.9511 Epoch 168/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2580e-04 - accuracy: 1.0000 - val_loss: 0.5106 - val_accuracy: 0.9510 Epoch 169/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2361e-04 - accuracy: 1.0000 - val_loss: 0.5109 - val_accuracy: 0.9509 Epoch 170/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2178e-04 - accuracy: 1.0000 - val_loss: 0.5129 - val_accuracy: 0.9510 Epoch 171/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1992e-04 - accuracy: 1.0000 - val_loss: 0.5140 - val_accuracy: 0.9512 Epoch 172/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1804e-04 - accuracy: 1.0000 - val_loss: 0.5131 - val_accuracy: 0.9510 Epoch 173/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1638e-04 - accuracy: 1.0000 - val_loss: 0.5149 - val_accuracy: 0.9508 Epoch 174/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1473e-04 - accuracy: 1.0000 - val_loss: 0.5154 - val_accuracy: 0.9510 Epoch 175/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1306e-04 - accuracy: 1.0000 - val_loss: 0.5170 - val_accuracy: 0.9507 Epoch 176/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1139e-04 - accuracy: 1.0000 - val_loss: 0.5175 - val_accuracy: 0.9511 Epoch 177/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0995e-04 - accuracy: 1.0000 - val_loss: 0.5182 - val_accuracy: 0.9509 Epoch 178/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0828e-04 - accuracy: 1.0000 - val_loss: 0.5192 - val_accuracy: 0.9513 Epoch 179/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0692e-04 - accuracy: 1.0000 - val_loss: 0.5191 - val_accuracy: 0.9510 Epoch 180/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0545e-04 - accuracy: 1.0000 - val_loss: 0.5209 - val_accuracy: 0.9509 Epoch 181/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0408e-04 - accuracy: 1.0000 - val_loss: 0.5210 - val_accuracy: 0.9510 Epoch 182/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0274e-04 - accuracy: 1.0000 - val_loss: 0.5227 - val_accuracy: 0.9510 Epoch 183/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0133e-04 - accuracy: 1.0000 - val_loss: 0.5239 - val_accuracy: 0.9512 Epoch 184/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0012e-04 - accuracy: 1.0000 - val_loss: 0.5240 - val_accuracy: 0.9506 Epoch 185/200 938/938 [==============================] - 3s 3ms/step - loss: 9.8911e-05 - accuracy: 1.0000 - val_loss: 0.5238 - val_accuracy: 0.9507 Epoch 186/200 938/938 [==============================] - 3s 3ms/step - loss: 9.7578e-05 - accuracy: 1.0000 - val_loss: 0.5242 - val_accuracy: 0.9510 Epoch 187/200 938/938 [==============================] - 3s 3ms/step - loss: 9.6527e-05 - accuracy: 1.0000 - val_loss: 0.5257 - val_accuracy: 0.9507 Epoch 188/200 938/938 [==============================] - 3s 3ms/step - loss: 9.5328e-05 - accuracy: 1.0000 - val_loss: 0.5261 - val_accuracy: 0.9508 Epoch 189/200 938/938 [==============================] - 3s 3ms/step - loss: 9.4206e-05 - accuracy: 1.0000 - val_loss: 0.5275 - val_accuracy: 0.9510 Epoch 190/200 938/938 [==============================] - 3s 3ms/step - loss: 9.3074e-05 - accuracy: 1.0000 - val_loss: 0.5273 - val_accuracy: 0.9511 Epoch 191/200 938/938 [==============================] - 3s 3ms/step - loss: 9.1907e-05 - accuracy: 1.0000 - val_loss: 0.5273 - val_accuracy: 0.9513 Epoch 192/200 938/938 [==============================] - 3s 3ms/step - loss: 9.0945e-05 - accuracy: 1.0000 - val_loss: 0.5291 - val_accuracy: 0.9510 Epoch 193/200 938/938 [==============================] - 3s 3ms/step - loss: 8.9916e-05 - accuracy: 1.0000 - val_loss: 0.5298 - val_accuracy: 0.9510 Epoch 194/200 938/938 [==============================] - 3s 3ms/step - loss: 8.8934e-05 - accuracy: 1.0000 - val_loss: 0.5303 - val_accuracy: 0.9507 Epoch 195/200 938/938 [==============================] - 3s 3ms/step - loss: 8.7953e-05 - accuracy: 1.0000 - val_loss: 0.5310 - val_accuracy: 0.9512 Epoch 196/200 938/938 [==============================] - 3s 3ms/step - loss: 8.7007e-05 - accuracy: 1.0000 - val_loss: 0.5313 - val_accuracy: 0.9514 Epoch 197/200 938/938 [==============================] - 3s 3ms/step - loss: 8.6052e-05 - accuracy: 1.0000 - val_loss: 0.5315 - val_accuracy: 0.9514 Epoch 198/200 938/938 [==============================] - 3s 3ms/step - loss: 8.5118e-05 - accuracy: 1.0000 - val_loss: 0.5327 - val_accuracy: 0.9511 Epoch 199/200 938/938 [==============================] - 3s 3ms/step - loss: 8.4178e-05 - accuracy: 1.0000 - val_loss: 0.5335 - val_accuracy: 0.9512 Epoch 200/200 938/938 [==============================] - 3s 3ms/step - loss: 8.3281e-05 - accuracy: 1.0000 - val_loss: 0.5352 - val_accuracy: 0.9510
(d) Activation function: ReLU; initialization: Xavier initializer¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.GlorotNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_11" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_11 (Flatten) (None, 784) 0 _________________________________________________________________ dense_66 (Dense) (None, 512) 401920 _________________________________________________________________ dense_67 (Dense) (None, 512) 262656 _________________________________________________________________ dense_68 (Dense) (None, 512) 262656 _________________________________________________________________ dense_69 (Dense) (None, 512) 262656 _________________________________________________________________ dense_70 (Dense) (None, 512) 262656 _________________________________________________________________ dense_71 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.SGD(learning_rate=.01) #learning_rate=1.0 for SGD
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
history3 = model.fit(X_train, y_train, batch_size=64, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 938/938 [==============================] - 3s 3ms/step - loss: 0.8977 - accuracy: 0.7730 - val_loss: 0.3496 - val_accuracy: 0.9001 Epoch 2/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2921 - accuracy: 0.9151 - val_loss: 0.2622 - val_accuracy: 0.9233 Epoch 3/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2230 - accuracy: 0.9346 - val_loss: 0.1998 - val_accuracy: 0.9398 Epoch 4/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1809 - accuracy: 0.9469 - val_loss: 0.1613 - val_accuracy: 0.9509 Epoch 5/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1529 - accuracy: 0.9557 - val_loss: 0.1406 - val_accuracy: 0.9578 Epoch 6/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1315 - accuracy: 0.9618 - val_loss: 0.1319 - val_accuracy: 0.9613 Epoch 7/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1152 - accuracy: 0.9667 - val_loss: 0.1193 - val_accuracy: 0.9653 Epoch 8/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1013 - accuracy: 0.9704 - val_loss: 0.1093 - val_accuracy: 0.9662 Epoch 9/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0897 - accuracy: 0.9738 - val_loss: 0.1294 - val_accuracy: 0.9598 Epoch 10/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0801 - accuracy: 0.9765 - val_loss: 0.1067 - val_accuracy: 0.9678 Epoch 11/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0714 - accuracy: 0.9793 - val_loss: 0.0953 - val_accuracy: 0.9714 Epoch 12/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0635 - accuracy: 0.9815 - val_loss: 0.1125 - val_accuracy: 0.9672 Epoch 13/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0564 - accuracy: 0.9839 - val_loss: 0.0842 - val_accuracy: 0.9742 Epoch 14/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0515 - accuracy: 0.9852 - val_loss: 0.0819 - val_accuracy: 0.9752 Epoch 15/200 938/938 [==============================] - 3s 4ms/step - loss: 0.0462 - accuracy: 0.9868 - val_loss: 0.0803 - val_accuracy: 0.9762 Epoch 16/200 938/938 [==============================] - 3s 4ms/step - loss: 0.0417 - accuracy: 0.9880 - val_loss: 0.0813 - val_accuracy: 0.9756 Epoch 17/200 938/938 [==============================] - 3s 4ms/step - loss: 0.0367 - accuracy: 0.9899 - val_loss: 0.0807 - val_accuracy: 0.9759 Epoch 18/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0330 - accuracy: 0.9912 - val_loss: 0.0797 - val_accuracy: 0.9762 Epoch 19/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0298 - accuracy: 0.9920 - val_loss: 0.0747 - val_accuracy: 0.9774 Epoch 20/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0264 - accuracy: 0.9931 - val_loss: 0.0776 - val_accuracy: 0.9773 Epoch 21/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0234 - accuracy: 0.9942 - val_loss: 0.0774 - val_accuracy: 0.9778 Epoch 22/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0207 - accuracy: 0.9952 - val_loss: 0.0761 - val_accuracy: 0.9779 Epoch 23/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0185 - accuracy: 0.9957 - val_loss: 0.0824 - val_accuracy: 0.9753 Epoch 24/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0165 - accuracy: 0.9964 - val_loss: 0.0769 - val_accuracy: 0.9779 Epoch 25/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0146 - accuracy: 0.9971 - val_loss: 0.0788 - val_accuracy: 0.9778 Epoch 26/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0128 - accuracy: 0.9974 - val_loss: 0.0762 - val_accuracy: 0.9792 Epoch 27/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0112 - accuracy: 0.9981 - val_loss: 0.0785 - val_accuracy: 0.9785 Epoch 28/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0101 - accuracy: 0.9983 - val_loss: 0.0858 - val_accuracy: 0.9772 Epoch 29/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0087 - accuracy: 0.9988 - val_loss: 0.0772 - val_accuracy: 0.9793 Epoch 30/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0075 - accuracy: 0.9991 - val_loss: 0.0820 - val_accuracy: 0.9774 Epoch 31/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0070 - accuracy: 0.9990 - val_loss: 0.0823 - val_accuracy: 0.9780 Epoch 32/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0061 - accuracy: 0.9994 - val_loss: 0.0783 - val_accuracy: 0.9793 Epoch 33/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0053 - accuracy: 0.9995 - val_loss: 0.0795 - val_accuracy: 0.9796 Epoch 34/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0048 - accuracy: 0.9996 - val_loss: 0.0842 - val_accuracy: 0.9788 Epoch 35/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0043 - accuracy: 0.9996 - val_loss: 0.0801 - val_accuracy: 0.9795 Epoch 36/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0037 - accuracy: 0.9998 - val_loss: 0.0807 - val_accuracy: 0.9794 Epoch 37/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0034 - accuracy: 0.9998 - val_loss: 0.0833 - val_accuracy: 0.9792 Epoch 38/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0030 - accuracy: 0.9998 - val_loss: 0.0837 - val_accuracy: 0.9791 Epoch 39/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0027 - accuracy: 0.9999 - val_loss: 0.0840 - val_accuracy: 0.9793 Epoch 40/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0025 - accuracy: 0.9999 - val_loss: 0.0850 - val_accuracy: 0.9786 Epoch 41/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0023 - accuracy: 0.9999 - val_loss: 0.0845 - val_accuracy: 0.9799 Epoch 42/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0021 - accuracy: 0.9999 - val_loss: 0.0865 - val_accuracy: 0.9787 Epoch 43/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0019 - accuracy: 1.0000 - val_loss: 0.0872 - val_accuracy: 0.9795 Epoch 44/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0018 - accuracy: 0.9999 - val_loss: 0.0882 - val_accuracy: 0.9793 Epoch 45/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0017 - accuracy: 1.0000 - val_loss: 0.0881 - val_accuracy: 0.9791 Epoch 46/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0015 - accuracy: 1.0000 - val_loss: 0.0883 - val_accuracy: 0.9798 Epoch 47/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0014 - accuracy: 1.0000 - val_loss: 0.0894 - val_accuracy: 0.9793 Epoch 48/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0013 - accuracy: 1.0000 - val_loss: 0.0900 - val_accuracy: 0.9795 Epoch 49/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0013 - accuracy: 1.0000 - val_loss: 0.0903 - val_accuracy: 0.9795 Epoch 50/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0012 - accuracy: 1.0000 - val_loss: 0.0905 - val_accuracy: 0.9795 Epoch 51/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0011 - accuracy: 1.0000 - val_loss: 0.0910 - val_accuracy: 0.9798 Epoch 52/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0011 - accuracy: 1.0000 - val_loss: 0.0919 - val_accuracy: 0.9795 Epoch 53/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0010 - accuracy: 1.0000 - val_loss: 0.0918 - val_accuracy: 0.9797 Epoch 54/200 938/938 [==============================] - 3s 3ms/step - loss: 9.6542e-04 - accuracy: 1.0000 - val_loss: 0.0926 - val_accuracy: 0.9794 Epoch 55/200 938/938 [==============================] - 3s 3ms/step - loss: 9.1859e-04 - accuracy: 1.0000 - val_loss: 0.0927 - val_accuracy: 0.9796 Epoch 56/200 938/938 [==============================] - 3s 3ms/step - loss: 8.9172e-04 - accuracy: 1.0000 - val_loss: 0.0940 - val_accuracy: 0.9792 Epoch 57/200 938/938 [==============================] - 3s 3ms/step - loss: 8.4012e-04 - accuracy: 1.0000 - val_loss: 0.0937 - val_accuracy: 0.9792 Epoch 58/200 938/938 [==============================] - 3s 3ms/step - loss: 8.0657e-04 - accuracy: 1.0000 - val_loss: 0.0942 - val_accuracy: 0.9793 Epoch 59/200 938/938 [==============================] - 3s 3ms/step - loss: 7.7955e-04 - accuracy: 1.0000 - val_loss: 0.0947 - val_accuracy: 0.9796 Epoch 60/200 938/938 [==============================] - 3s 3ms/step - loss: 7.4536e-04 - accuracy: 1.0000 - val_loss: 0.0949 - val_accuracy: 0.9794 Epoch 61/200 938/938 [==============================] - 3s 3ms/step - loss: 7.1695e-04 - accuracy: 1.0000 - val_loss: 0.0954 - val_accuracy: 0.9795 Epoch 62/200 938/938 [==============================] - 3s 3ms/step - loss: 6.9276e-04 - accuracy: 1.0000 - val_loss: 0.0959 - val_accuracy: 0.9793 Epoch 63/200 938/938 [==============================] - 3s 3ms/step - loss: 6.6305e-04 - accuracy: 1.0000 - val_loss: 0.0965 - val_accuracy: 0.9792 Epoch 64/200 938/938 [==============================] - 3s 3ms/step - loss: 6.4406e-04 - accuracy: 1.0000 - val_loss: 0.0962 - val_accuracy: 0.9796 Epoch 65/200 938/938 [==============================] - 3s 3ms/step - loss: 6.1780e-04 - accuracy: 1.0000 - val_loss: 0.0971 - val_accuracy: 0.9793 Epoch 66/200 938/938 [==============================] - 3s 3ms/step - loss: 6.0194e-04 - accuracy: 1.0000 - val_loss: 0.0971 - val_accuracy: 0.9793 Epoch 67/200 938/938 [==============================] - 3s 3ms/step - loss: 5.8031e-04 - accuracy: 1.0000 - val_loss: 0.0972 - val_accuracy: 0.9795 Epoch 68/200 938/938 [==============================] - 3s 3ms/step - loss: 5.6361e-04 - accuracy: 1.0000 - val_loss: 0.0975 - val_accuracy: 0.9794 Epoch 69/200 938/938 [==============================] - 3s 3ms/step - loss: 5.4437e-04 - accuracy: 1.0000 - val_loss: 0.0984 - val_accuracy: 0.9795 Epoch 70/200 938/938 [==============================] - 3s 3ms/step - loss: 5.2660e-04 - accuracy: 1.0000 - val_loss: 0.0988 - val_accuracy: 0.9796 Epoch 71/200 938/938 [==============================] - 3s 3ms/step - loss: 5.1449e-04 - accuracy: 1.0000 - val_loss: 0.0988 - val_accuracy: 0.9792 Epoch 72/200 938/938 [==============================] - 3s 3ms/step - loss: 4.9714e-04 - accuracy: 1.0000 - val_loss: 0.0991 - val_accuracy: 0.9797 Epoch 73/200 938/938 [==============================] - 3s 3ms/step - loss: 4.8457e-04 - accuracy: 1.0000 - val_loss: 0.0989 - val_accuracy: 0.9792 Epoch 74/200 938/938 [==============================] - 3s 3ms/step - loss: 4.7051e-04 - accuracy: 1.0000 - val_loss: 0.1000 - val_accuracy: 0.9796 Epoch 75/200 938/938 [==============================] - 3s 3ms/step - loss: 4.6025e-04 - accuracy: 1.0000 - val_loss: 0.1000 - val_accuracy: 0.9798 Epoch 76/200 938/938 [==============================] - 3s 3ms/step - loss: 4.4789e-04 - accuracy: 1.0000 - val_loss: 0.1003 - val_accuracy: 0.9796 Epoch 77/200 938/938 [==============================] - 3s 3ms/step - loss: 4.3601e-04 - accuracy: 1.0000 - val_loss: 0.1006 - val_accuracy: 0.9796 Epoch 78/200 938/938 [==============================] - 3s 3ms/step - loss: 4.2445e-04 - accuracy: 1.0000 - val_loss: 0.1009 - val_accuracy: 0.9793 Epoch 79/200 938/938 [==============================] - 3s 3ms/step - loss: 4.1331e-04 - accuracy: 1.0000 - val_loss: 0.1016 - val_accuracy: 0.9797 Epoch 80/200 938/938 [==============================] - 3s 3ms/step - loss: 4.0416e-04 - accuracy: 1.0000 - val_loss: 0.1016 - val_accuracy: 0.9799 Epoch 81/200 938/938 [==============================] - 3s 3ms/step - loss: 3.9688e-04 - accuracy: 1.0000 - val_loss: 0.1015 - val_accuracy: 0.9791 Epoch 82/200 938/938 [==============================] - 3s 3ms/step - loss: 3.8537e-04 - accuracy: 1.0000 - val_loss: 0.1019 - val_accuracy: 0.9794 Epoch 83/200 938/938 [==============================] - 3s 3ms/step - loss: 3.7603e-04 - accuracy: 1.0000 - val_loss: 0.1022 - val_accuracy: 0.9797 Epoch 84/200 938/938 [==============================] - 3s 3ms/step - loss: 3.6579e-04 - accuracy: 1.0000 - val_loss: 0.1026 - val_accuracy: 0.9796 Epoch 85/200 938/938 [==============================] - 3s 3ms/step - loss: 3.5945e-04 - accuracy: 1.0000 - val_loss: 0.1027 - val_accuracy: 0.9797 Epoch 86/200 938/938 [==============================] - 3s 3ms/step - loss: 3.5085e-04 - accuracy: 1.0000 - val_loss: 0.1033 - val_accuracy: 0.9793 Epoch 87/200 938/938 [==============================] - 3s 3ms/step - loss: 3.4380e-04 - accuracy: 1.0000 - val_loss: 0.1033 - val_accuracy: 0.9793 Epoch 88/200 938/938 [==============================] - 3s 3ms/step - loss: 3.3564e-04 - accuracy: 1.0000 - val_loss: 0.1035 - val_accuracy: 0.9791 Epoch 89/200 938/938 [==============================] - 3s 3ms/step - loss: 3.2963e-04 - accuracy: 1.0000 - val_loss: 0.1039 - val_accuracy: 0.9793 Epoch 90/200 938/938 [==============================] - 3s 3ms/step - loss: 3.2194e-04 - accuracy: 1.0000 - val_loss: 0.1040 - val_accuracy: 0.9795 Epoch 91/200 938/938 [==============================] - 3s 3ms/step - loss: 3.1592e-04 - accuracy: 1.0000 - val_loss: 0.1043 - val_accuracy: 0.9794 Epoch 92/200 938/938 [==============================] - 3s 3ms/step - loss: 3.0831e-04 - accuracy: 1.0000 - val_loss: 0.1045 - val_accuracy: 0.9796 Epoch 93/200 938/938 [==============================] - 3s 3ms/step - loss: 3.0449e-04 - accuracy: 1.0000 - val_loss: 0.1048 - val_accuracy: 0.9795 Epoch 94/200 938/938 [==============================] - 3s 3ms/step - loss: 2.9821e-04 - accuracy: 1.0000 - val_loss: 0.1049 - val_accuracy: 0.9796 Epoch 95/200 938/938 [==============================] - 3s 3ms/step - loss: 2.9174e-04 - accuracy: 1.0000 - val_loss: 0.1049 - val_accuracy: 0.9793 Epoch 96/200 938/938 [==============================] - 3s 3ms/step - loss: 2.8701e-04 - accuracy: 1.0000 - val_loss: 0.1052 - val_accuracy: 0.9795 Epoch 97/200 938/938 [==============================] - 3s 3ms/step - loss: 2.8113e-04 - accuracy: 1.0000 - val_loss: 0.1060 - val_accuracy: 0.9794 Epoch 98/200 938/938 [==============================] - 3s 3ms/step - loss: 2.7680e-04 - accuracy: 1.0000 - val_loss: 0.1054 - val_accuracy: 0.9795 Epoch 99/200 938/938 [==============================] - 3s 3ms/step - loss: 2.7150e-04 - accuracy: 1.0000 - val_loss: 0.1060 - val_accuracy: 0.9795 Epoch 100/200 938/938 [==============================] - 3s 3ms/step - loss: 2.6644e-04 - accuracy: 1.0000 - val_loss: 0.1057 - val_accuracy: 0.9795 Epoch 101/200 938/938 [==============================] - 3s 3ms/step - loss: 2.6226e-04 - accuracy: 1.0000 - val_loss: 0.1063 - val_accuracy: 0.9795 Epoch 102/200 938/938 [==============================] - 3s 3ms/step - loss: 2.5747e-04 - accuracy: 1.0000 - val_loss: 0.1067 - val_accuracy: 0.9795 Epoch 103/200 938/938 [==============================] - 3s 3ms/step - loss: 2.5215e-04 - accuracy: 1.0000 - val_loss: 0.1069 - val_accuracy: 0.9792 Epoch 104/200 938/938 [==============================] - 3s 3ms/step - loss: 2.4906e-04 - accuracy: 1.0000 - val_loss: 0.1067 - val_accuracy: 0.9793 Epoch 105/200 938/938 [==============================] - 3s 3ms/step - loss: 2.4488e-04 - accuracy: 1.0000 - val_loss: 0.1072 - val_accuracy: 0.9794 Epoch 106/200 938/938 [==============================] - 3s 3ms/step - loss: 2.4007e-04 - accuracy: 1.0000 - val_loss: 0.1073 - val_accuracy: 0.9793 Epoch 107/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3686e-04 - accuracy: 1.0000 - val_loss: 0.1075 - val_accuracy: 0.9794 Epoch 108/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3300e-04 - accuracy: 1.0000 - val_loss: 0.1077 - val_accuracy: 0.9795 Epoch 109/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2931e-04 - accuracy: 1.0000 - val_loss: 0.1076 - val_accuracy: 0.9795 Epoch 110/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2575e-04 - accuracy: 1.0000 - val_loss: 0.1079 - val_accuracy: 0.9793 Epoch 111/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2165e-04 - accuracy: 1.0000 - val_loss: 0.1079 - val_accuracy: 0.9795 Epoch 112/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1828e-04 - accuracy: 1.0000 - val_loss: 0.1080 - val_accuracy: 0.9795 Epoch 113/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1533e-04 - accuracy: 1.0000 - val_loss: 0.1086 - val_accuracy: 0.9797 Epoch 114/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1235e-04 - accuracy: 1.0000 - val_loss: 0.1086 - val_accuracy: 0.9795 Epoch 115/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0896e-04 - accuracy: 1.0000 - val_loss: 0.1086 - val_accuracy: 0.9794 Epoch 116/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0651e-04 - accuracy: 1.0000 - val_loss: 0.1090 - val_accuracy: 0.9795 Epoch 117/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0269e-04 - accuracy: 1.0000 - val_loss: 0.1092 - val_accuracy: 0.9794 Epoch 118/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0018e-04 - accuracy: 1.0000 - val_loss: 0.1092 - val_accuracy: 0.9794 Epoch 119/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9724e-04 - accuracy: 1.0000 - val_loss: 0.1093 - val_accuracy: 0.9794 Epoch 120/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9458e-04 - accuracy: 1.0000 - val_loss: 0.1094 - val_accuracy: 0.9793 Epoch 121/200 938/938 [==============================] - 4s 4ms/step - loss: 1.9208e-04 - accuracy: 1.0000 - val_loss: 0.1100 - val_accuracy: 0.9796 Epoch 122/200 938/938 [==============================] - 3s 4ms/step - loss: 1.8917e-04 - accuracy: 1.0000 - val_loss: 0.1098 - val_accuracy: 0.9794 Epoch 123/200 938/938 [==============================] - 3s 4ms/step - loss: 1.8650e-04 - accuracy: 1.0000 - val_loss: 0.1101 - val_accuracy: 0.9794 Epoch 124/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8429e-04 - accuracy: 1.0000 - val_loss: 0.1102 - val_accuracy: 0.9793 Epoch 125/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8174e-04 - accuracy: 1.0000 - val_loss: 0.1102 - val_accuracy: 0.9796 Epoch 126/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7905e-04 - accuracy: 1.0000 - val_loss: 0.1107 - val_accuracy: 0.9795 Epoch 127/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7691e-04 - accuracy: 1.0000 - val_loss: 0.1108 - val_accuracy: 0.9797 Epoch 128/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7483e-04 - accuracy: 1.0000 - val_loss: 0.1108 - val_accuracy: 0.9793 Epoch 129/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7300e-04 - accuracy: 1.0000 - val_loss: 0.1110 - val_accuracy: 0.9795 Epoch 130/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7062e-04 - accuracy: 1.0000 - val_loss: 0.1110 - val_accuracy: 0.9793 Epoch 131/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6841e-04 - accuracy: 1.0000 - val_loss: 0.1112 - val_accuracy: 0.9795 Epoch 132/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6598e-04 - accuracy: 1.0000 - val_loss: 0.1115 - val_accuracy: 0.9795 Epoch 133/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6433e-04 - accuracy: 1.0000 - val_loss: 0.1115 - val_accuracy: 0.9795 Epoch 134/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6219e-04 - accuracy: 1.0000 - val_loss: 0.1116 - val_accuracy: 0.9795 Epoch 135/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6012e-04 - accuracy: 1.0000 - val_loss: 0.1116 - val_accuracy: 0.9796 Epoch 136/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5838e-04 - accuracy: 1.0000 - val_loss: 0.1117 - val_accuracy: 0.9796 Epoch 137/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5624e-04 - accuracy: 1.0000 - val_loss: 0.1122 - val_accuracy: 0.9795 Epoch 138/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5393e-04 - accuracy: 1.0000 - val_loss: 0.1121 - val_accuracy: 0.9796 Epoch 139/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5289e-04 - accuracy: 1.0000 - val_loss: 0.1124 - val_accuracy: 0.9795 Epoch 140/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5069e-04 - accuracy: 1.0000 - val_loss: 0.1124 - val_accuracy: 0.9792 Epoch 141/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4943e-04 - accuracy: 1.0000 - val_loss: 0.1127 - val_accuracy: 0.9795 Epoch 142/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4753e-04 - accuracy: 1.0000 - val_loss: 0.1129 - val_accuracy: 0.9794 Epoch 143/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4609e-04 - accuracy: 1.0000 - val_loss: 0.1127 - val_accuracy: 0.9795 Epoch 144/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4431e-04 - accuracy: 1.0000 - val_loss: 0.1130 - val_accuracy: 0.9792 Epoch 145/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4291e-04 - accuracy: 1.0000 - val_loss: 0.1131 - val_accuracy: 0.9793 Epoch 146/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4122e-04 - accuracy: 1.0000 - val_loss: 0.1132 - val_accuracy: 0.9795 Epoch 147/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3981e-04 - accuracy: 1.0000 - val_loss: 0.1135 - val_accuracy: 0.9794 Epoch 148/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3830e-04 - accuracy: 1.0000 - val_loss: 0.1135 - val_accuracy: 0.9795 Epoch 149/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3680e-04 - accuracy: 1.0000 - val_loss: 0.1136 - val_accuracy: 0.9794 Epoch 150/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3523e-04 - accuracy: 1.0000 - val_loss: 0.1137 - val_accuracy: 0.9796 Epoch 151/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3412e-04 - accuracy: 1.0000 - val_loss: 0.1139 - val_accuracy: 0.9795 Epoch 152/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3264e-04 - accuracy: 1.0000 - val_loss: 0.1138 - val_accuracy: 0.9795 Epoch 153/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3086e-04 - accuracy: 1.0000 - val_loss: 0.1140 - val_accuracy: 0.9792 Epoch 154/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2994e-04 - accuracy: 1.0000 - val_loss: 0.1142 - val_accuracy: 0.9793 Epoch 155/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2864e-04 - accuracy: 1.0000 - val_loss: 0.1144 - val_accuracy: 0.9793 Epoch 156/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2743e-04 - accuracy: 1.0000 - val_loss: 0.1144 - val_accuracy: 0.9793 Epoch 157/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2604e-04 - accuracy: 1.0000 - val_loss: 0.1146 - val_accuracy: 0.9796 Epoch 158/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2463e-04 - accuracy: 1.0000 - val_loss: 0.1146 - val_accuracy: 0.9796 Epoch 159/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2339e-04 - accuracy: 1.0000 - val_loss: 0.1148 - val_accuracy: 0.9794 Epoch 160/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2237e-04 - accuracy: 1.0000 - val_loss: 0.1149 - val_accuracy: 0.9796 Epoch 161/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2100e-04 - accuracy: 1.0000 - val_loss: 0.1151 - val_accuracy: 0.9795 Epoch 162/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1989e-04 - accuracy: 1.0000 - val_loss: 0.1151 - val_accuracy: 0.9794 Epoch 163/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1858e-04 - accuracy: 1.0000 - val_loss: 0.1150 - val_accuracy: 0.9793 Epoch 164/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1787e-04 - accuracy: 1.0000 - val_loss: 0.1154 - val_accuracy: 0.9795 Epoch 165/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1644e-04 - accuracy: 1.0000 - val_loss: 0.1155 - val_accuracy: 0.9795 Epoch 166/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1555e-04 - accuracy: 1.0000 - val_loss: 0.1156 - val_accuracy: 0.9795 Epoch 167/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1433e-04 - accuracy: 1.0000 - val_loss: 0.1160 - val_accuracy: 0.9797 Epoch 168/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1352e-04 - accuracy: 1.0000 - val_loss: 0.1157 - val_accuracy: 0.9793 Epoch 169/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1241e-04 - accuracy: 1.0000 - val_loss: 0.1161 - val_accuracy: 0.9797 Epoch 170/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1142e-04 - accuracy: 1.0000 - val_loss: 0.1159 - val_accuracy: 0.9792 Epoch 171/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1043e-04 - accuracy: 1.0000 - val_loss: 0.1163 - val_accuracy: 0.9794 Epoch 172/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0937e-04 - accuracy: 1.0000 - val_loss: 0.1161 - val_accuracy: 0.9795 Epoch 173/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0840e-04 - accuracy: 1.0000 - val_loss: 0.1163 - val_accuracy: 0.9794 Epoch 174/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0733e-04 - accuracy: 1.0000 - val_loss: 0.1163 - val_accuracy: 0.9795 Epoch 175/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0651e-04 - accuracy: 1.0000 - val_loss: 0.1165 - val_accuracy: 0.9795 Epoch 176/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0554e-04 - accuracy: 1.0000 - val_loss: 0.1166 - val_accuracy: 0.9795 Epoch 177/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0467e-04 - accuracy: 1.0000 - val_loss: 0.1167 - val_accuracy: 0.9794 Epoch 178/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0368e-04 - accuracy: 1.0000 - val_loss: 0.1166 - val_accuracy: 0.9792 Epoch 179/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0295e-04 - accuracy: 1.0000 - val_loss: 0.1169 - val_accuracy: 0.9794 Epoch 180/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0207e-04 - accuracy: 1.0000 - val_loss: 0.1171 - val_accuracy: 0.9794 Epoch 181/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0123e-04 - accuracy: 1.0000 - val_loss: 0.1172 - val_accuracy: 0.9795 Epoch 182/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0023e-04 - accuracy: 1.0000 - val_loss: 0.1172 - val_accuracy: 0.9795 Epoch 183/200 938/938 [==============================] - 3s 3ms/step - loss: 9.9611e-05 - accuracy: 1.0000 - val_loss: 0.1172 - val_accuracy: 0.9793 Epoch 184/200 938/938 [==============================] - 3s 3ms/step - loss: 9.8657e-05 - accuracy: 1.0000 - val_loss: 0.1176 - val_accuracy: 0.9795 Epoch 185/200 938/938 [==============================] - 3s 3ms/step - loss: 9.8070e-05 - accuracy: 1.0000 - val_loss: 0.1174 - val_accuracy: 0.9793 Epoch 186/200 938/938 [==============================] - 3s 3ms/step - loss: 9.7171e-05 - accuracy: 1.0000 - val_loss: 0.1174 - val_accuracy: 0.9793 Epoch 187/200 938/938 [==============================] - 3s 3ms/step - loss: 9.6257e-05 - accuracy: 1.0000 - val_loss: 0.1178 - val_accuracy: 0.9795 Epoch 188/200 938/938 [==============================] - 3s 3ms/step - loss: 9.5757e-05 - accuracy: 1.0000 - val_loss: 0.1176 - val_accuracy: 0.9792 Epoch 189/200 938/938 [==============================] - 3s 3ms/step - loss: 9.4810e-05 - accuracy: 1.0000 - val_loss: 0.1179 - val_accuracy: 0.9793 Epoch 190/200 938/938 [==============================] - 3s 3ms/step - loss: 9.4068e-05 - accuracy: 1.0000 - val_loss: 0.1179 - val_accuracy: 0.9794 Epoch 191/200 938/938 [==============================] - 3s 3ms/step - loss: 9.3322e-05 - accuracy: 1.0000 - val_loss: 0.1180 - val_accuracy: 0.9793 Epoch 192/200 938/938 [==============================] - 3s 3ms/step - loss: 9.2460e-05 - accuracy: 1.0000 - val_loss: 0.1181 - val_accuracy: 0.9793 Epoch 193/200 938/938 [==============================] - 3s 3ms/step - loss: 9.2000e-05 - accuracy: 1.0000 - val_loss: 0.1182 - val_accuracy: 0.9794 Epoch 194/200 938/938 [==============================] - 3s 3ms/step - loss: 9.1224e-05 - accuracy: 1.0000 - val_loss: 0.1182 - val_accuracy: 0.9794 Epoch 195/200 938/938 [==============================] - 3s 3ms/step - loss: 9.0363e-05 - accuracy: 1.0000 - val_loss: 0.1185 - val_accuracy: 0.9794 Epoch 196/200 938/938 [==============================] - 3s 3ms/step - loss: 8.9874e-05 - accuracy: 1.0000 - val_loss: 0.1185 - val_accuracy: 0.9794 Epoch 197/200 938/938 [==============================] - 3s 3ms/step - loss: 8.9103e-05 - accuracy: 1.0000 - val_loss: 0.1185 - val_accuracy: 0.9794 Epoch 198/200 938/938 [==============================] - 3s 3ms/step - loss: 8.8384e-05 - accuracy: 1.0000 - val_loss: 0.1188 - val_accuracy: 0.9795 Epoch 199/200 938/938 [==============================] - 3s 3ms/step - loss: 8.7737e-05 - accuracy: 1.0000 - val_loss: 0.1190 - val_accuracy: 0.9796 Epoch 200/200 938/938 [==============================] - 3s 3ms/step - loss: 8.7152e-05 - accuracy: 1.0000 - val_loss: 0.1187 - val_accuracy: 0.9792
(e) Activation function: ReLU; initialization: Kaiming He’s initializer¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.HeNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_12" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_12 (Flatten) (None, 784) 0 _________________________________________________________________ dense_72 (Dense) (None, 512) 401920 _________________________________________________________________ dense_73 (Dense) (None, 512) 262656 _________________________________________________________________ dense_74 (Dense) (None, 512) 262656 _________________________________________________________________ dense_75 (Dense) (None, 512) 262656 _________________________________________________________________ dense_76 (Dense) (None, 512) 262656 _________________________________________________________________ dense_77 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.SGD(learning_rate=.01) #learning_rate=1.0 for SGD
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
history4 = model.fit(X_train, y_train, batch_size=64, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 938/938 [==============================] - 3s 3ms/step - loss: 0.4928 - accuracy: 0.8651 - val_loss: 0.2472 - val_accuracy: 0.9233 Epoch 2/200 938/938 [==============================] - 3s 3ms/step - loss: 0.2065 - accuracy: 0.9394 - val_loss: 0.1962 - val_accuracy: 0.9422 Epoch 3/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1548 - accuracy: 0.9543 - val_loss: 0.1572 - val_accuracy: 0.9541 Epoch 4/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1245 - accuracy: 0.9632 - val_loss: 0.1322 - val_accuracy: 0.9588 Epoch 5/200 938/938 [==============================] - 3s 3ms/step - loss: 0.1033 - accuracy: 0.9694 - val_loss: 0.1237 - val_accuracy: 0.9627 Epoch 6/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0876 - accuracy: 0.9741 - val_loss: 0.1034 - val_accuracy: 0.9708 Epoch 7/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0748 - accuracy: 0.9779 - val_loss: 0.1010 - val_accuracy: 0.9693 Epoch 8/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0650 - accuracy: 0.9811 - val_loss: 0.0875 - val_accuracy: 0.9742 Epoch 9/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0569 - accuracy: 0.9834 - val_loss: 0.1152 - val_accuracy: 0.9633 Epoch 10/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0490 - accuracy: 0.9857 - val_loss: 0.0868 - val_accuracy: 0.9736 Epoch 11/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0427 - accuracy: 0.9878 - val_loss: 0.0795 - val_accuracy: 0.9761 Epoch 12/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0370 - accuracy: 0.9893 - val_loss: 0.0856 - val_accuracy: 0.9741 Epoch 13/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0318 - accuracy: 0.9917 - val_loss: 0.1016 - val_accuracy: 0.9693 Epoch 14/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0276 - accuracy: 0.9929 - val_loss: 0.0821 - val_accuracy: 0.9757 Epoch 15/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0232 - accuracy: 0.9944 - val_loss: 0.0863 - val_accuracy: 0.9739 Epoch 16/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0207 - accuracy: 0.9951 - val_loss: 0.0793 - val_accuracy: 0.9766 Epoch 17/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0179 - accuracy: 0.9961 - val_loss: 0.0839 - val_accuracy: 0.9746 Epoch 18/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0154 - accuracy: 0.9969 - val_loss: 0.0767 - val_accuracy: 0.9779 Epoch 19/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0130 - accuracy: 0.9978 - val_loss: 0.0772 - val_accuracy: 0.9775 Epoch 20/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0116 - accuracy: 0.9981 - val_loss: 0.0749 - val_accuracy: 0.9778 Epoch 21/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0097 - accuracy: 0.9985 - val_loss: 0.0777 - val_accuracy: 0.9769 Epoch 22/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0084 - accuracy: 0.9990 - val_loss: 0.0773 - val_accuracy: 0.9787 Epoch 23/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0073 - accuracy: 0.9992 - val_loss: 0.0784 - val_accuracy: 0.9793 Epoch 24/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0060 - accuracy: 0.9996 - val_loss: 0.0781 - val_accuracy: 0.9786 Epoch 25/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0053 - accuracy: 0.9995 - val_loss: 0.0767 - val_accuracy: 0.9785 Epoch 26/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0046 - accuracy: 0.9997 - val_loss: 0.0823 - val_accuracy: 0.9788 Epoch 27/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0041 - accuracy: 0.9998 - val_loss: 0.0784 - val_accuracy: 0.9796 Epoch 28/200 938/938 [==============================] - 3s 4ms/step - loss: 0.0037 - accuracy: 0.9999 - val_loss: 0.0826 - val_accuracy: 0.9781 Epoch 29/200 938/938 [==============================] - 3s 4ms/step - loss: 0.0032 - accuracy: 0.9999 - val_loss: 0.0825 - val_accuracy: 0.9786 Epoch 30/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0029 - accuracy: 0.9999 - val_loss: 0.0806 - val_accuracy: 0.9793 Epoch 31/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0026 - accuracy: 0.9999 - val_loss: 0.0817 - val_accuracy: 0.9790 Epoch 32/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0024 - accuracy: 1.0000 - val_loss: 0.0821 - val_accuracy: 0.9793 Epoch 33/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0021 - accuracy: 1.0000 - val_loss: 0.0823 - val_accuracy: 0.9788 Epoch 34/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0020 - accuracy: 1.0000 - val_loss: 0.0830 - val_accuracy: 0.9788 Epoch 35/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0019 - accuracy: 1.0000 - val_loss: 0.0840 - val_accuracy: 0.9788 Epoch 36/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0017 - accuracy: 1.0000 - val_loss: 0.0856 - val_accuracy: 0.9794 Epoch 37/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0016 - accuracy: 1.0000 - val_loss: 0.0855 - val_accuracy: 0.9791 Epoch 38/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0015 - accuracy: 1.0000 - val_loss: 0.0856 - val_accuracy: 0.9791 Epoch 39/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0014 - accuracy: 1.0000 - val_loss: 0.0863 - val_accuracy: 0.9791 Epoch 40/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0013 - accuracy: 1.0000 - val_loss: 0.0868 - val_accuracy: 0.9792 Epoch 41/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0013 - accuracy: 1.0000 - val_loss: 0.0872 - val_accuracy: 0.9784 Epoch 42/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0012 - accuracy: 1.0000 - val_loss: 0.0876 - val_accuracy: 0.9784 Epoch 43/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0011 - accuracy: 1.0000 - val_loss: 0.0878 - val_accuracy: 0.9792 Epoch 44/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0011 - accuracy: 1.0000 - val_loss: 0.0878 - val_accuracy: 0.9797 Epoch 45/200 938/938 [==============================] - 3s 3ms/step - loss: 0.0010 - accuracy: 1.0000 - val_loss: 0.0885 - val_accuracy: 0.9795 Epoch 46/200 938/938 [==============================] - 3s 3ms/step - loss: 9.9241e-04 - accuracy: 1.0000 - val_loss: 0.0884 - val_accuracy: 0.9794 Epoch 47/200 938/938 [==============================] - 3s 3ms/step - loss: 9.5525e-04 - accuracy: 1.0000 - val_loss: 0.0888 - val_accuracy: 0.9790 Epoch 48/200 938/938 [==============================] - 3s 3ms/step - loss: 9.1258e-04 - accuracy: 1.0000 - val_loss: 0.0898 - val_accuracy: 0.9789 Epoch 49/200 938/938 [==============================] - 3s 3ms/step - loss: 8.7487e-04 - accuracy: 1.0000 - val_loss: 0.0895 - val_accuracy: 0.9792 Epoch 50/200 938/938 [==============================] - 3s 3ms/step - loss: 8.4040e-04 - accuracy: 1.0000 - val_loss: 0.0894 - val_accuracy: 0.9791 Epoch 51/200 938/938 [==============================] - 3s 3ms/step - loss: 8.1012e-04 - accuracy: 1.0000 - val_loss: 0.0904 - val_accuracy: 0.9791 Epoch 52/200 938/938 [==============================] - 3s 3ms/step - loss: 7.7838e-04 - accuracy: 1.0000 - val_loss: 0.0901 - val_accuracy: 0.9791 Epoch 53/200 938/938 [==============================] - 3s 3ms/step - loss: 7.5134e-04 - accuracy: 1.0000 - val_loss: 0.0915 - val_accuracy: 0.9795 Epoch 54/200 938/938 [==============================] - 3s 3ms/step - loss: 7.2590e-04 - accuracy: 1.0000 - val_loss: 0.0909 - val_accuracy: 0.9792 Epoch 55/200 938/938 [==============================] - 3s 3ms/step - loss: 7.0316e-04 - accuracy: 1.0000 - val_loss: 0.0924 - val_accuracy: 0.9785 Epoch 56/200 938/938 [==============================] - 3s 3ms/step - loss: 6.8198e-04 - accuracy: 1.0000 - val_loss: 0.0918 - val_accuracy: 0.9792 Epoch 57/200 938/938 [==============================] - 3s 3ms/step - loss: 6.5640e-04 - accuracy: 1.0000 - val_loss: 0.0920 - val_accuracy: 0.9793 Epoch 58/200 938/938 [==============================] - 3s 3ms/step - loss: 6.3500e-04 - accuracy: 1.0000 - val_loss: 0.0927 - val_accuracy: 0.9790 Epoch 59/200 938/938 [==============================] - 3s 3ms/step - loss: 6.1509e-04 - accuracy: 1.0000 - val_loss: 0.0927 - val_accuracy: 0.9790 Epoch 60/200 938/938 [==============================] - 3s 3ms/step - loss: 5.9790e-04 - accuracy: 1.0000 - val_loss: 0.0932 - val_accuracy: 0.9793 Epoch 61/200 938/938 [==============================] - 3s 3ms/step - loss: 5.7882e-04 - accuracy: 1.0000 - val_loss: 0.0930 - val_accuracy: 0.9791 Epoch 62/200 938/938 [==============================] - 3s 3ms/step - loss: 5.6287e-04 - accuracy: 1.0000 - val_loss: 0.0934 - val_accuracy: 0.9793 Epoch 63/200 938/938 [==============================] - 3s 3ms/step - loss: 5.4409e-04 - accuracy: 1.0000 - val_loss: 0.0939 - val_accuracy: 0.9793 Epoch 64/200 938/938 [==============================] - 3s 3ms/step - loss: 5.3308e-04 - accuracy: 1.0000 - val_loss: 0.0942 - val_accuracy: 0.9795 Epoch 65/200 938/938 [==============================] - 3s 3ms/step - loss: 5.1492e-04 - accuracy: 1.0000 - val_loss: 0.0944 - val_accuracy: 0.9798 Epoch 66/200 938/938 [==============================] - 3s 3ms/step - loss: 5.0429e-04 - accuracy: 1.0000 - val_loss: 0.0948 - val_accuracy: 0.9798 Epoch 67/200 938/938 [==============================] - 3s 3ms/step - loss: 4.9105e-04 - accuracy: 1.0000 - val_loss: 0.0945 - val_accuracy: 0.9794 Epoch 68/200 938/938 [==============================] - 3s 3ms/step - loss: 4.7805e-04 - accuracy: 1.0000 - val_loss: 0.0952 - val_accuracy: 0.9792 Epoch 69/200 938/938 [==============================] - 3s 3ms/step - loss: 4.6546e-04 - accuracy: 1.0000 - val_loss: 0.0952 - val_accuracy: 0.9791 Epoch 70/200 938/938 [==============================] - 3s 3ms/step - loss: 4.5494e-04 - accuracy: 1.0000 - val_loss: 0.0948 - val_accuracy: 0.9795 Epoch 71/200 938/938 [==============================] - 3s 3ms/step - loss: 4.4527e-04 - accuracy: 1.0000 - val_loss: 0.0956 - val_accuracy: 0.9792 Epoch 72/200 938/938 [==============================] - 3s 3ms/step - loss: 4.3459e-04 - accuracy: 1.0000 - val_loss: 0.0954 - val_accuracy: 0.9794 Epoch 73/200 938/938 [==============================] - 3s 3ms/step - loss: 4.2344e-04 - accuracy: 1.0000 - val_loss: 0.0958 - val_accuracy: 0.9791 Epoch 74/200 938/938 [==============================] - 3s 3ms/step - loss: 4.1433e-04 - accuracy: 1.0000 - val_loss: 0.0961 - val_accuracy: 0.9793 Epoch 75/200 938/938 [==============================] - 3s 3ms/step - loss: 4.0497e-04 - accuracy: 1.0000 - val_loss: 0.0963 - val_accuracy: 0.9794 Epoch 76/200 938/938 [==============================] - 3s 3ms/step - loss: 3.9504e-04 - accuracy: 1.0000 - val_loss: 0.0967 - val_accuracy: 0.9795 Epoch 77/200 938/938 [==============================] - 3s 3ms/step - loss: 3.8727e-04 - accuracy: 1.0000 - val_loss: 0.0969 - val_accuracy: 0.9797 Epoch 78/200 938/938 [==============================] - 3s 3ms/step - loss: 3.7946e-04 - accuracy: 1.0000 - val_loss: 0.0972 - val_accuracy: 0.9794 Epoch 79/200 938/938 [==============================] - 3s 3ms/step - loss: 3.7156e-04 - accuracy: 1.0000 - val_loss: 0.0973 - val_accuracy: 0.9792 Epoch 80/200 938/938 [==============================] - 3s 3ms/step - loss: 3.6358e-04 - accuracy: 1.0000 - val_loss: 0.0972 - val_accuracy: 0.9793 Epoch 81/200 938/938 [==============================] - 3s 3ms/step - loss: 3.5582e-04 - accuracy: 1.0000 - val_loss: 0.0975 - val_accuracy: 0.9789 Epoch 82/200 938/938 [==============================] - 3s 3ms/step - loss: 3.4802e-04 - accuracy: 1.0000 - val_loss: 0.0982 - val_accuracy: 0.9788 Epoch 83/200 938/938 [==============================] - 3s 3ms/step - loss: 3.4214e-04 - accuracy: 1.0000 - val_loss: 0.0982 - val_accuracy: 0.9793 Epoch 84/200 938/938 [==============================] - 3s 3ms/step - loss: 3.3511e-04 - accuracy: 1.0000 - val_loss: 0.0985 - val_accuracy: 0.9792 Epoch 85/200 938/938 [==============================] - 3s 3ms/step - loss: 3.2912e-04 - accuracy: 1.0000 - val_loss: 0.0985 - val_accuracy: 0.9798 Epoch 86/200 938/938 [==============================] - 3s 3ms/step - loss: 3.2387e-04 - accuracy: 1.0000 - val_loss: 0.0984 - val_accuracy: 0.9792 Epoch 87/200 938/938 [==============================] - 3s 3ms/step - loss: 3.1683e-04 - accuracy: 1.0000 - val_loss: 0.0991 - val_accuracy: 0.9790 Epoch 88/200 938/938 [==============================] - 3s 3ms/step - loss: 3.1116e-04 - accuracy: 1.0000 - val_loss: 0.0990 - val_accuracy: 0.9797 Epoch 89/200 938/938 [==============================] - 3s 3ms/step - loss: 3.0520e-04 - accuracy: 1.0000 - val_loss: 0.0995 - val_accuracy: 0.9793 Epoch 90/200 938/938 [==============================] - 3s 3ms/step - loss: 3.0045e-04 - accuracy: 1.0000 - val_loss: 0.0992 - val_accuracy: 0.9793 Epoch 91/200 938/938 [==============================] - 3s 3ms/step - loss: 2.9531e-04 - accuracy: 1.0000 - val_loss: 0.0992 - val_accuracy: 0.9796 Epoch 92/200 938/938 [==============================] - 3s 3ms/step - loss: 2.9000e-04 - accuracy: 1.0000 - val_loss: 0.0996 - val_accuracy: 0.9793 Epoch 93/200 938/938 [==============================] - 3s 3ms/step - loss: 2.8475e-04 - accuracy: 1.0000 - val_loss: 0.0999 - val_accuracy: 0.9795 Epoch 94/200 938/938 [==============================] - 3s 3ms/step - loss: 2.8072e-04 - accuracy: 1.0000 - val_loss: 0.0996 - val_accuracy: 0.9793 Epoch 95/200 938/938 [==============================] - 3s 3ms/step - loss: 2.7546e-04 - accuracy: 1.0000 - val_loss: 0.1002 - val_accuracy: 0.9792 Epoch 96/200 938/938 [==============================] - 3s 3ms/step - loss: 2.7144e-04 - accuracy: 1.0000 - val_loss: 0.1000 - val_accuracy: 0.9795 Epoch 97/200 938/938 [==============================] - 3s 3ms/step - loss: 2.6583e-04 - accuracy: 1.0000 - val_loss: 0.1003 - val_accuracy: 0.9791 Epoch 98/200 938/938 [==============================] - 3s 3ms/step - loss: 2.6156e-04 - accuracy: 1.0000 - val_loss: 0.1009 - val_accuracy: 0.9796 Epoch 99/200 938/938 [==============================] - 3s 3ms/step - loss: 2.5903e-04 - accuracy: 1.0000 - val_loss: 0.1008 - val_accuracy: 0.9795 Epoch 100/200 938/938 [==============================] - 3s 3ms/step - loss: 2.5408e-04 - accuracy: 1.0000 - val_loss: 0.1010 - val_accuracy: 0.9793 Epoch 101/200 938/938 [==============================] - 3s 3ms/step - loss: 2.5049e-04 - accuracy: 1.0000 - val_loss: 0.1013 - val_accuracy: 0.9795 Epoch 102/200 938/938 [==============================] - 3s 3ms/step - loss: 2.4697e-04 - accuracy: 1.0000 - val_loss: 0.1012 - val_accuracy: 0.9796 Epoch 103/200 938/938 [==============================] - 3s 3ms/step - loss: 2.4326e-04 - accuracy: 1.0000 - val_loss: 0.1014 - val_accuracy: 0.9796 Epoch 104/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3966e-04 - accuracy: 1.0000 - val_loss: 0.1015 - val_accuracy: 0.9795 Epoch 105/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3599e-04 - accuracy: 1.0000 - val_loss: 0.1019 - val_accuracy: 0.9793 Epoch 106/200 938/938 [==============================] - 3s 3ms/step - loss: 2.3225e-04 - accuracy: 1.0000 - val_loss: 0.1018 - val_accuracy: 0.9795 Epoch 107/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2901e-04 - accuracy: 1.0000 - val_loss: 0.1017 - val_accuracy: 0.9794 Epoch 108/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2594e-04 - accuracy: 1.0000 - val_loss: 0.1019 - val_accuracy: 0.9794 Epoch 109/200 938/938 [==============================] - 3s 3ms/step - loss: 2.2282e-04 - accuracy: 1.0000 - val_loss: 0.1023 - val_accuracy: 0.9794 Epoch 110/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1955e-04 - accuracy: 1.0000 - val_loss: 0.1024 - val_accuracy: 0.9793 Epoch 111/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1644e-04 - accuracy: 1.0000 - val_loss: 0.1025 - val_accuracy: 0.9795 Epoch 112/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1294e-04 - accuracy: 1.0000 - val_loss: 0.1025 - val_accuracy: 0.9794 Epoch 113/200 938/938 [==============================] - 3s 3ms/step - loss: 2.1068e-04 - accuracy: 1.0000 - val_loss: 0.1026 - val_accuracy: 0.9793 Epoch 114/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0776e-04 - accuracy: 1.0000 - val_loss: 0.1028 - val_accuracy: 0.9795 Epoch 115/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0502e-04 - accuracy: 1.0000 - val_loss: 0.1027 - val_accuracy: 0.9794 Epoch 116/200 938/938 [==============================] - 3s 3ms/step - loss: 2.0213e-04 - accuracy: 1.0000 - val_loss: 0.1031 - val_accuracy: 0.9795 Epoch 117/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9974e-04 - accuracy: 1.0000 - val_loss: 0.1029 - val_accuracy: 0.9793 Epoch 118/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9762e-04 - accuracy: 1.0000 - val_loss: 0.1035 - val_accuracy: 0.9794 Epoch 119/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9475e-04 - accuracy: 1.0000 - val_loss: 0.1033 - val_accuracy: 0.9794 Epoch 120/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9237e-04 - accuracy: 1.0000 - val_loss: 0.1035 - val_accuracy: 0.9796 Epoch 121/200 938/938 [==============================] - 3s 3ms/step - loss: 1.9000e-04 - accuracy: 1.0000 - val_loss: 0.1036 - val_accuracy: 0.9796 Epoch 122/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8774e-04 - accuracy: 1.0000 - val_loss: 0.1039 - val_accuracy: 0.9795 Epoch 123/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8559e-04 - accuracy: 1.0000 - val_loss: 0.1040 - val_accuracy: 0.9795 Epoch 124/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8307e-04 - accuracy: 1.0000 - val_loss: 0.1040 - val_accuracy: 0.9794 Epoch 125/200 938/938 [==============================] - 3s 3ms/step - loss: 1.8084e-04 - accuracy: 1.0000 - val_loss: 0.1040 - val_accuracy: 0.9794 Epoch 126/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7878e-04 - accuracy: 1.0000 - val_loss: 0.1044 - val_accuracy: 0.9795 Epoch 127/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7669e-04 - accuracy: 1.0000 - val_loss: 0.1046 - val_accuracy: 0.9795 Epoch 128/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7464e-04 - accuracy: 1.0000 - val_loss: 0.1045 - val_accuracy: 0.9796 Epoch 129/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7250e-04 - accuracy: 1.0000 - val_loss: 0.1045 - val_accuracy: 0.9796 Epoch 130/200 938/938 [==============================] - 3s 3ms/step - loss: 1.7091e-04 - accuracy: 1.0000 - val_loss: 0.1047 - val_accuracy: 0.9796 Epoch 131/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6856e-04 - accuracy: 1.0000 - val_loss: 0.1049 - val_accuracy: 0.9797 Epoch 132/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6671e-04 - accuracy: 1.0000 - val_loss: 0.1049 - val_accuracy: 0.9794 Epoch 133/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6482e-04 - accuracy: 1.0000 - val_loss: 0.1052 - val_accuracy: 0.9796 Epoch 134/200 938/938 [==============================] - 3s 4ms/step - loss: 1.6301e-04 - accuracy: 1.0000 - val_loss: 0.1053 - val_accuracy: 0.9794 Epoch 135/200 938/938 [==============================] - 3s 3ms/step - loss: 1.6122e-04 - accuracy: 1.0000 - val_loss: 0.1052 - val_accuracy: 0.9795 Epoch 136/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5952e-04 - accuracy: 1.0000 - val_loss: 0.1053 - val_accuracy: 0.9794 Epoch 137/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5799e-04 - accuracy: 1.0000 - val_loss: 0.1057 - val_accuracy: 0.9795 Epoch 138/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5617e-04 - accuracy: 1.0000 - val_loss: 0.1053 - val_accuracy: 0.9794 Epoch 139/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5437e-04 - accuracy: 1.0000 - val_loss: 0.1055 - val_accuracy: 0.9794 Epoch 140/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5275e-04 - accuracy: 1.0000 - val_loss: 0.1058 - val_accuracy: 0.9796 Epoch 141/200 938/938 [==============================] - 3s 3ms/step - loss: 1.5144e-04 - accuracy: 1.0000 - val_loss: 0.1057 - val_accuracy: 0.9796 Epoch 142/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4973e-04 - accuracy: 1.0000 - val_loss: 0.1058 - val_accuracy: 0.9794 Epoch 143/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4831e-04 - accuracy: 1.0000 - val_loss: 0.1061 - val_accuracy: 0.9796 Epoch 144/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4638e-04 - accuracy: 1.0000 - val_loss: 0.1062 - val_accuracy: 0.9796 Epoch 145/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4542e-04 - accuracy: 1.0000 - val_loss: 0.1059 - val_accuracy: 0.9796 Epoch 146/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4384e-04 - accuracy: 1.0000 - val_loss: 0.1063 - val_accuracy: 0.9797 Epoch 147/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4243e-04 - accuracy: 1.0000 - val_loss: 0.1066 - val_accuracy: 0.9796 Epoch 148/200 938/938 [==============================] - 3s 3ms/step - loss: 1.4094e-04 - accuracy: 1.0000 - val_loss: 0.1066 - val_accuracy: 0.9795 Epoch 149/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3968e-04 - accuracy: 1.0000 - val_loss: 0.1066 - val_accuracy: 0.9794 Epoch 150/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3827e-04 - accuracy: 1.0000 - val_loss: 0.1067 - val_accuracy: 0.9797 Epoch 151/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3694e-04 - accuracy: 1.0000 - val_loss: 0.1068 - val_accuracy: 0.9795 Epoch 152/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3567e-04 - accuracy: 1.0000 - val_loss: 0.1069 - val_accuracy: 0.9796 Epoch 153/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3401e-04 - accuracy: 1.0000 - val_loss: 0.1075 - val_accuracy: 0.9796 Epoch 154/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3297e-04 - accuracy: 1.0000 - val_loss: 0.1071 - val_accuracy: 0.9797 Epoch 155/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3184e-04 - accuracy: 1.0000 - val_loss: 0.1071 - val_accuracy: 0.9798 Epoch 156/200 938/938 [==============================] - 3s 3ms/step - loss: 1.3063e-04 - accuracy: 1.0000 - val_loss: 0.1072 - val_accuracy: 0.9797 Epoch 157/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2945e-04 - accuracy: 1.0000 - val_loss: 0.1074 - val_accuracy: 0.9795 Epoch 158/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2814e-04 - accuracy: 1.0000 - val_loss: 0.1075 - val_accuracy: 0.9798 Epoch 159/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2713e-04 - accuracy: 1.0000 - val_loss: 0.1075 - val_accuracy: 0.9797 Epoch 160/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2603e-04 - accuracy: 1.0000 - val_loss: 0.1075 - val_accuracy: 0.9796 Epoch 161/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2470e-04 - accuracy: 1.0000 - val_loss: 0.1078 - val_accuracy: 0.9797 Epoch 162/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2373e-04 - accuracy: 1.0000 - val_loss: 0.1078 - val_accuracy: 0.9795 Epoch 163/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2273e-04 - accuracy: 1.0000 - val_loss: 0.1081 - val_accuracy: 0.9796 Epoch 164/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2161e-04 - accuracy: 1.0000 - val_loss: 0.1081 - val_accuracy: 0.9796 Epoch 165/200 938/938 [==============================] - 3s 3ms/step - loss: 1.2041e-04 - accuracy: 1.0000 - val_loss: 0.1079 - val_accuracy: 0.9795 Epoch 166/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1927e-04 - accuracy: 1.0000 - val_loss: 0.1082 - val_accuracy: 0.9796 Epoch 167/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1838e-04 - accuracy: 1.0000 - val_loss: 0.1082 - val_accuracy: 0.9794 Epoch 168/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1745e-04 - accuracy: 1.0000 - val_loss: 0.1084 - val_accuracy: 0.9797 Epoch 169/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1648e-04 - accuracy: 1.0000 - val_loss: 0.1085 - val_accuracy: 0.9797 Epoch 170/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1547e-04 - accuracy: 1.0000 - val_loss: 0.1085 - val_accuracy: 0.9797 Epoch 171/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1463e-04 - accuracy: 1.0000 - val_loss: 0.1085 - val_accuracy: 0.9798 Epoch 172/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1365e-04 - accuracy: 1.0000 - val_loss: 0.1086 - val_accuracy: 0.9796 Epoch 173/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1267e-04 - accuracy: 1.0000 - val_loss: 0.1089 - val_accuracy: 0.9796 Epoch 174/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1162e-04 - accuracy: 1.0000 - val_loss: 0.1088 - val_accuracy: 0.9797 Epoch 175/200 938/938 [==============================] - 3s 3ms/step - loss: 1.1072e-04 - accuracy: 1.0000 - val_loss: 0.1090 - val_accuracy: 0.9796 Epoch 176/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0988e-04 - accuracy: 1.0000 - val_loss: 0.1090 - val_accuracy: 0.9796 Epoch 177/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0872e-04 - accuracy: 1.0000 - val_loss: 0.1091 - val_accuracy: 0.9797 Epoch 178/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0818e-04 - accuracy: 1.0000 - val_loss: 0.1093 - val_accuracy: 0.9797 Epoch 179/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0736e-04 - accuracy: 1.0000 - val_loss: 0.1091 - val_accuracy: 0.9795 Epoch 180/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0653e-04 - accuracy: 1.0000 - val_loss: 0.1092 - val_accuracy: 0.9795 Epoch 181/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0560e-04 - accuracy: 1.0000 - val_loss: 0.1092 - val_accuracy: 0.9795 Epoch 182/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0469e-04 - accuracy: 1.0000 - val_loss: 0.1096 - val_accuracy: 0.9796 Epoch 183/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0395e-04 - accuracy: 1.0000 - val_loss: 0.1095 - val_accuracy: 0.9796 Epoch 184/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0317e-04 - accuracy: 1.0000 - val_loss: 0.1098 - val_accuracy: 0.9797 Epoch 185/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0233e-04 - accuracy: 1.0000 - val_loss: 0.1097 - val_accuracy: 0.9796 Epoch 186/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0171e-04 - accuracy: 1.0000 - val_loss: 0.1098 - val_accuracy: 0.9797 Epoch 187/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0081e-04 - accuracy: 1.0000 - val_loss: 0.1099 - val_accuracy: 0.9797 Epoch 188/200 938/938 [==============================] - 3s 3ms/step - loss: 1.0012e-04 - accuracy: 1.0000 - val_loss: 0.1100 - val_accuracy: 0.9796 Epoch 189/200 938/938 [==============================] - 3s 3ms/step - loss: 9.9270e-05 - accuracy: 1.0000 - val_loss: 0.1102 - val_accuracy: 0.9794 Epoch 190/200 938/938 [==============================] - 3s 3ms/step - loss: 9.8603e-05 - accuracy: 1.0000 - val_loss: 0.1103 - val_accuracy: 0.9799 Epoch 191/200 938/938 [==============================] - 3s 3ms/step - loss: 9.7870e-05 - accuracy: 1.0000 - val_loss: 0.1102 - val_accuracy: 0.9796 Epoch 192/200 938/938 [==============================] - 3s 3ms/step - loss: 9.7116e-05 - accuracy: 1.0000 - val_loss: 0.1103 - val_accuracy: 0.9795 Epoch 193/200 938/938 [==============================] - 3s 3ms/step - loss: 9.6447e-05 - accuracy: 1.0000 - val_loss: 0.1105 - val_accuracy: 0.9797 Epoch 194/200 938/938 [==============================] - 3s 3ms/step - loss: 9.5707e-05 - accuracy: 1.0000 - val_loss: 0.1102 - val_accuracy: 0.9796 Epoch 195/200 938/938 [==============================] - 3s 3ms/step - loss: 9.4960e-05 - accuracy: 1.0000 - val_loss: 0.1104 - val_accuracy: 0.9797 Epoch 196/200 938/938 [==============================] - 3s 3ms/step - loss: 9.4395e-05 - accuracy: 1.0000 - val_loss: 0.1106 - val_accuracy: 0.9797 Epoch 197/200 938/938 [==============================] - 3s 3ms/step - loss: 9.3783e-05 - accuracy: 1.0000 - val_loss: 0.1107 - val_accuracy: 0.9797 Epoch 198/200 938/938 [==============================] - 3s 3ms/step - loss: 9.3070e-05 - accuracy: 1.0000 - val_loss: 0.1108 - val_accuracy: 0.9797 Epoch 199/200 938/938 [==============================] - 3s 3ms/step - loss: 9.2503e-05 - accuracy: 1.0000 - val_loss: 0.1108 - val_accuracy: 0.9795 Epoch 200/200 938/938 [==============================] - 3s 3ms/step - loss: 9.1673e-05 - accuracy: 1.0000 - val_loss: 0.1110 - val_accuracy: 0.9798
Visualize SGD Results¶
In [ ]:
#loss_train = history.history['accuracy']
test_acc = history.history['val_accuracy'][0:50]
test_acc1 = history1.history['val_accuracy'][0:50]
test_acc2 = history2.history['val_accuracy'][0:50]
test_acc3 = history3.history['val_accuracy'][0:50]
test_acc4 = history4.history['val_accuracy'][0:50]
epochs = range(0,50)
plt.figure(figsize=(20,10))
plt.plot(epochs, test_acc, 'r', label='Logistic: Normal')
plt.plot(epochs, test_acc1, 'r', label='Logistic: Xavier', linewidth=3)
plt.plot(epochs, test_acc2, 'b', label='ReLU: Normal')
plt.plot(epochs, test_acc3, 'b', label='ReLU: Xavier', linewidth=3)
plt.plot(epochs, test_acc4, 'black', label='ReLU: He')
plt.title('SGD')
plt.xlabel('Epoch')
plt.ylabel('Test Accuracy %')
plt.legend()
plt.show()
In [ ]:
#loss_train = history.history['accuracy']
test_acc = history.history['val_accuracy'][51:]
test_acc1 = history1.history['val_accuracy'][51:]
test_acc2 = history2.history['val_accuracy'][51:]
test_acc3 = history3.history['val_accuracy'][51:]
test_acc4 = history4.history['val_accuracy'][51:]
epochs = range(51,200)
plt.figure(figsize=(20,10))
plt.plot(epochs, test_acc, 'r', label='Logistic: Normal')
plt.plot(epochs, test_acc1, 'r', label='Logistic: Xavier', linewidth=3)
plt.plot(epochs, test_acc2, 'b', label='ReLU: Normal')
plt.plot(epochs, test_acc3, 'b', label='ReLU: Xavier', linewidth=3)
plt.plot(epochs, test_acc4, 'black', label='ReLU: He')
plt.title('SGD')
plt.xlabel('Epoch')
plt.ylabel('Test Accuracy %')
plt.legend()
plt.show()
Adam¶
(a) Activation function: the logistic sigmoid function; initialization: random numbers gen-erated from the normal distribution (μ = 0, σ = 0.01)¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.01, seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
#tf.keras.layers.Dense(512,kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_3" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_3 (Flatten) (None, 784) 0 _________________________________________________________________ dense_18 (Dense) (None, 512) 401920 _________________________________________________________________ dense_19 (Dense) (None, 512) 262656 _________________________________________________________________ dense_20 (Dense) (None, 512) 262656 _________________________________________________________________ dense_21 (Dense) (None, 512) 262656 _________________________________________________________________ dense_22 (Dense) (None, 512) 262656 _________________________________________________________________ dense_23 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam() #learning_rate=1.0 for SGD
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
historya = model.fit(X_train, y_train, batch_size=128, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 469/469 [==============================] - 2s 4ms/step - loss: 2.1356 - accuracy: 0.1633 - val_loss: 1.9725 - val_accuracy: 0.2173 Epoch 2/200 469/469 [==============================] - 2s 3ms/step - loss: 1.3498 - accuracy: 0.4763 - val_loss: 0.7668 - val_accuracy: 0.7515 Epoch 3/200 469/469 [==============================] - 2s 3ms/step - loss: 0.4624 - accuracy: 0.8750 - val_loss: 0.3131 - val_accuracy: 0.9143 Epoch 4/200 469/469 [==============================] - 2s 3ms/step - loss: 0.2746 - accuracy: 0.9271 - val_loss: 0.2702 - val_accuracy: 0.9296 Epoch 5/200 469/469 [==============================] - 2s 3ms/step - loss: 0.2111 - accuracy: 0.9431 - val_loss: 0.2163 - val_accuracy: 0.9406 Epoch 6/200 469/469 [==============================] - 2s 3ms/step - loss: 0.1697 - accuracy: 0.9540 - val_loss: 0.1626 - val_accuracy: 0.9569 Epoch 7/200 469/469 [==============================] - 2s 3ms/step - loss: 0.1388 - accuracy: 0.9618 - val_loss: 0.1526 - val_accuracy: 0.9587 Epoch 8/200 469/469 [==============================] - 2s 3ms/step - loss: 0.1154 - accuracy: 0.9675 - val_loss: 0.1289 - val_accuracy: 0.9647 Epoch 9/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0967 - accuracy: 0.9729 - val_loss: 0.1449 - val_accuracy: 0.9599 Epoch 10/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0841 - accuracy: 0.9757 - val_loss: 0.1163 - val_accuracy: 0.9674 Epoch 11/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0725 - accuracy: 0.9794 - val_loss: 0.1121 - val_accuracy: 0.9697 Epoch 12/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0592 - accuracy: 0.9831 - val_loss: 0.1169 - val_accuracy: 0.9676 Epoch 13/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0544 - accuracy: 0.9839 - val_loss: 0.1156 - val_accuracy: 0.9708 Epoch 14/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0474 - accuracy: 0.9862 - val_loss: 0.1006 - val_accuracy: 0.9728 Epoch 15/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0409 - accuracy: 0.9883 - val_loss: 0.1169 - val_accuracy: 0.9698 Epoch 16/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0350 - accuracy: 0.9896 - val_loss: 0.1027 - val_accuracy: 0.9737 Epoch 17/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0363 - accuracy: 0.9892 - val_loss: 0.1063 - val_accuracy: 0.9750 Epoch 18/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0284 - accuracy: 0.9914 - val_loss: 0.1031 - val_accuracy: 0.9766 Epoch 19/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0235 - accuracy: 0.9935 - val_loss: 0.1276 - val_accuracy: 0.9708 Epoch 20/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0233 - accuracy: 0.9933 - val_loss: 0.1445 - val_accuracy: 0.9687 Epoch 21/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0217 - accuracy: 0.9937 - val_loss: 0.1194 - val_accuracy: 0.9750 Epoch 22/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0223 - accuracy: 0.9933 - val_loss: 0.1114 - val_accuracy: 0.9761 Epoch 23/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0168 - accuracy: 0.9951 - val_loss: 0.1060 - val_accuracy: 0.9753 Epoch 24/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9958 - val_loss: 0.1274 - val_accuracy: 0.9753 Epoch 25/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0161 - accuracy: 0.9952 - val_loss: 0.1360 - val_accuracy: 0.9726 Epoch 26/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9961 - val_loss: 0.1209 - val_accuracy: 0.9762 Epoch 27/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0149 - accuracy: 0.9955 - val_loss: 0.1295 - val_accuracy: 0.9747 Epoch 28/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0120 - accuracy: 0.9965 - val_loss: 0.1184 - val_accuracy: 0.9758 Epoch 29/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0112 - accuracy: 0.9967 - val_loss: 0.1222 - val_accuracy: 0.9756 Epoch 30/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0115 - accuracy: 0.9965 - val_loss: 0.1199 - val_accuracy: 0.9787 Epoch 31/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0102 - accuracy: 0.9970 - val_loss: 0.1112 - val_accuracy: 0.9800 Epoch 32/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0094 - accuracy: 0.9974 - val_loss: 0.1135 - val_accuracy: 0.9789 Epoch 33/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0088 - accuracy: 0.9972 - val_loss: 0.1202 - val_accuracy: 0.9762 Epoch 34/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0110 - accuracy: 0.9968 - val_loss: 0.1163 - val_accuracy: 0.9783 Epoch 35/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0084 - accuracy: 0.9974 - val_loss: 0.1254 - val_accuracy: 0.9769 Epoch 36/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0080 - accuracy: 0.9976 - val_loss: 0.1225 - val_accuracy: 0.9774 Epoch 37/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0078 - accuracy: 0.9975 - val_loss: 0.1295 - val_accuracy: 0.9778 Epoch 38/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0054 - accuracy: 0.9985 - val_loss: 0.1286 - val_accuracy: 0.9783 Epoch 39/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0077 - accuracy: 0.9977 - val_loss: 0.1160 - val_accuracy: 0.9815 Epoch 40/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0072 - accuracy: 0.9980 - val_loss: 0.1272 - val_accuracy: 0.9763 Epoch 41/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0071 - accuracy: 0.9979 - val_loss: 0.1195 - val_accuracy: 0.9802 Epoch 42/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0069 - accuracy: 0.9981 - val_loss: 0.1175 - val_accuracy: 0.9793 Epoch 43/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0079 - accuracy: 0.9976 - val_loss: 0.1202 - val_accuracy: 0.9792 Epoch 44/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0057 - accuracy: 0.9985 - val_loss: 0.1221 - val_accuracy: 0.9799 Epoch 45/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0036 - accuracy: 0.9991 - val_loss: 0.1281 - val_accuracy: 0.9793 Epoch 46/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0091 - accuracy: 0.9973 - val_loss: 0.1166 - val_accuracy: 0.9798 Epoch 47/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0037 - accuracy: 0.9990 - val_loss: 0.1245 - val_accuracy: 0.9793 Epoch 48/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0055 - accuracy: 0.9983 - val_loss: 0.1297 - val_accuracy: 0.9782 Epoch 49/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0056 - accuracy: 0.9983 - val_loss: 0.1304 - val_accuracy: 0.9781 Epoch 50/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0043 - accuracy: 0.9987 - val_loss: 0.1240 - val_accuracy: 0.9811 Epoch 51/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0090 - accuracy: 0.9974 - val_loss: 0.1090 - val_accuracy: 0.9821 Epoch 52/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0043 - accuracy: 0.9987 - val_loss: 0.1240 - val_accuracy: 0.9784 Epoch 53/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0047 - accuracy: 0.9986 - val_loss: 0.1153 - val_accuracy: 0.9819 Epoch 54/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0048 - accuracy: 0.9986 - val_loss: 0.1208 - val_accuracy: 0.9803 Epoch 55/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0035 - accuracy: 0.9989 - val_loss: 0.1220 - val_accuracy: 0.9805 Epoch 56/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0055 - accuracy: 0.9984 - val_loss: 0.1303 - val_accuracy: 0.9781 Epoch 57/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0049 - accuracy: 0.9985 - val_loss: 0.1516 - val_accuracy: 0.9765 Epoch 58/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0059 - accuracy: 0.9981 - val_loss: 0.1347 - val_accuracy: 0.9761 Epoch 59/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0015 - accuracy: 0.9996 - val_loss: 0.1199 - val_accuracy: 0.9823 Epoch 60/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9989 - val_loss: 0.1304 - val_accuracy: 0.9807 Epoch 61/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9985 - val_loss: 0.1187 - val_accuracy: 0.9814 Epoch 62/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0038 - accuracy: 0.9988 - val_loss: 0.1225 - val_accuracy: 0.9806 Epoch 63/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0026 - accuracy: 0.9993 - val_loss: 0.1288 - val_accuracy: 0.9820 Epoch 64/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0051 - accuracy: 0.9987 - val_loss: 0.1309 - val_accuracy: 0.9803 Epoch 65/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0042 - accuracy: 0.9988 - val_loss: 0.1178 - val_accuracy: 0.9815 Epoch 66/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0027 - accuracy: 0.9993 - val_loss: 0.1339 - val_accuracy: 0.9796 Epoch 67/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0048 - accuracy: 0.9985 - val_loss: 0.1222 - val_accuracy: 0.9800 Epoch 68/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0035 - accuracy: 0.9988 - val_loss: 0.1488 - val_accuracy: 0.9775 Epoch 69/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0044 - accuracy: 0.9987 - val_loss: 0.1166 - val_accuracy: 0.9816 Epoch 70/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0016 - accuracy: 0.9995 - val_loss: 0.1232 - val_accuracy: 0.9823 Epoch 71/200 469/469 [==============================] - 2s 3ms/step - loss: 6.5848e-04 - accuracy: 0.9999 - val_loss: 0.1323 - val_accuracy: 0.9793 Epoch 72/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0036 - accuracy: 0.9990 - val_loss: 0.1404 - val_accuracy: 0.9794 Epoch 73/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0068 - accuracy: 0.9981 - val_loss: 0.1481 - val_accuracy: 0.9761 Epoch 74/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0034 - accuracy: 0.9990 - val_loss: 0.1367 - val_accuracy: 0.9791 Epoch 75/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0037 - accuracy: 0.9989 - val_loss: 0.1354 - val_accuracy: 0.9793 Epoch 76/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0045 - accuracy: 0.9988 - val_loss: 0.1237 - val_accuracy: 0.9814 Epoch 77/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0023 - accuracy: 0.9993 - val_loss: 0.1307 - val_accuracy: 0.9808 Epoch 78/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0020 - accuracy: 0.9995 - val_loss: 0.1322 - val_accuracy: 0.9819 Epoch 79/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0055 - accuracy: 0.9984 - val_loss: 0.1461 - val_accuracy: 0.9764 Epoch 80/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0026 - accuracy: 0.9993 - val_loss: 0.1160 - val_accuracy: 0.9820 Epoch 81/200 469/469 [==============================] - 2s 4ms/step - loss: 6.8951e-04 - accuracy: 0.9998 - val_loss: 0.1241 - val_accuracy: 0.9823 Epoch 82/200 469/469 [==============================] - 2s 4ms/step - loss: 8.9305e-04 - accuracy: 0.9998 - val_loss: 0.1241 - val_accuracy: 0.9830 Epoch 83/200 469/469 [==============================] - 2s 3ms/step - loss: 1.8648e-04 - accuracy: 1.0000 - val_loss: 0.1324 - val_accuracy: 0.9826 Epoch 84/200 469/469 [==============================] - 2s 3ms/step - loss: 1.4409e-04 - accuracy: 0.9999 - val_loss: 0.1354 - val_accuracy: 0.9827 Epoch 85/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0025 - accuracy: 0.9994 - val_loss: 0.1421 - val_accuracy: 0.9795 Epoch 86/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0136 - accuracy: 0.9968 - val_loss: 0.1262 - val_accuracy: 0.9809 Epoch 87/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0010 - accuracy: 0.9997 - val_loss: 0.1125 - val_accuracy: 0.9842 Epoch 88/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0017 - accuracy: 0.9995 - val_loss: 0.1248 - val_accuracy: 0.9827 Epoch 89/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0065 - accuracy: 0.9980 - val_loss: 0.1174 - val_accuracy: 0.9802 Epoch 90/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0019 - accuracy: 0.9995 - val_loss: 0.1279 - val_accuracy: 0.9813 Epoch 91/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0021 - accuracy: 0.9995 - val_loss: 0.1247 - val_accuracy: 0.9825 Epoch 92/200 469/469 [==============================] - 2s 3ms/step - loss: 1.8251e-04 - accuracy: 1.0000 - val_loss: 0.1281 - val_accuracy: 0.9826 Epoch 93/200 469/469 [==============================] - 2s 3ms/step - loss: 2.4101e-04 - accuracy: 1.0000 - val_loss: 0.1292 - val_accuracy: 0.9824 Epoch 94/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2869e-04 - accuracy: 1.0000 - val_loss: 0.1314 - val_accuracy: 0.9831 Epoch 95/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0085 - accuracy: 0.9978 - val_loss: 0.1287 - val_accuracy: 0.9803 Epoch 96/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0050 - accuracy: 0.9988 - val_loss: 0.1450 - val_accuracy: 0.9777 Epoch 97/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0046 - accuracy: 0.9987 - val_loss: 0.1247 - val_accuracy: 0.9818 Epoch 98/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0017 - accuracy: 0.9995 - val_loss: 0.1317 - val_accuracy: 0.9821 Epoch 99/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0020 - accuracy: 0.9994 - val_loss: 0.1510 - val_accuracy: 0.9801 Epoch 100/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0021 - accuracy: 0.9995 - val_loss: 0.1840 - val_accuracy: 0.9747 Epoch 101/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0040 - accuracy: 0.9988 - val_loss: 0.1273 - val_accuracy: 0.9805 Epoch 102/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0012 - accuracy: 0.9997 - val_loss: 0.1291 - val_accuracy: 0.9816 Epoch 103/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0010 - accuracy: 0.9997 - val_loss: 0.1629 - val_accuracy: 0.9793 Epoch 104/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0030 - accuracy: 0.9992 - val_loss: 0.1206 - val_accuracy: 0.9832 Epoch 105/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0027 - accuracy: 0.9993 - val_loss: 0.1239 - val_accuracy: 0.9829 Epoch 106/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0014 - accuracy: 0.9997 - val_loss: 0.1462 - val_accuracy: 0.9809 Epoch 107/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0040 - accuracy: 0.9989 - val_loss: 0.1268 - val_accuracy: 0.9814 Epoch 108/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0027 - accuracy: 0.9992 - val_loss: 0.1222 - val_accuracy: 0.9832 Epoch 109/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0028 - accuracy: 0.9992 - val_loss: 0.1354 - val_accuracy: 0.9807 Epoch 110/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0011 - accuracy: 0.9997 - val_loss: 0.1306 - val_accuracy: 0.9832 Epoch 111/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0030 - accuracy: 0.9993 - val_loss: 0.1226 - val_accuracy: 0.9805 Epoch 112/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0033 - accuracy: 0.9992 - val_loss: 0.1681 - val_accuracy: 0.9723 Epoch 113/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0023 - accuracy: 0.9993 - val_loss: 0.1334 - val_accuracy: 0.9820 Epoch 114/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0022 - accuracy: 0.9994 - val_loss: 0.1232 - val_accuracy: 0.9818 Epoch 115/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0026 - accuracy: 0.9994 - val_loss: 0.1342 - val_accuracy: 0.9802 Epoch 116/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0027 - accuracy: 0.9993 - val_loss: 0.1305 - val_accuracy: 0.9811 Epoch 117/200 469/469 [==============================] - 2s 4ms/step - loss: 3.8401e-04 - accuracy: 0.9999 - val_loss: 0.1292 - val_accuracy: 0.9820 Epoch 118/200 469/469 [==============================] - 2s 4ms/step - loss: 4.3346e-05 - accuracy: 1.0000 - val_loss: 0.1359 - val_accuracy: 0.9816 Epoch 119/200 469/469 [==============================] - 2s 4ms/step - loss: 1.7511e-05 - accuracy: 1.0000 - val_loss: 0.1386 - val_accuracy: 0.9818 Epoch 120/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2479e-05 - accuracy: 1.0000 - val_loss: 0.1415 - val_accuracy: 0.9819 Epoch 121/200 469/469 [==============================] - 2s 3ms/step - loss: 9.3661e-06 - accuracy: 1.0000 - val_loss: 0.1443 - val_accuracy: 0.9821 Epoch 122/200 469/469 [==============================] - 2s 3ms/step - loss: 7.3577e-06 - accuracy: 1.0000 - val_loss: 0.1472 - val_accuracy: 0.9822 Epoch 123/200 469/469 [==============================] - 2s 3ms/step - loss: 5.7041e-06 - accuracy: 1.0000 - val_loss: 0.1501 - val_accuracy: 0.9822 Epoch 124/200 469/469 [==============================] - 2s 3ms/step - loss: 4.4694e-06 - accuracy: 1.0000 - val_loss: 0.1525 - val_accuracy: 0.9822 Epoch 125/200 469/469 [==============================] - 2s 4ms/step - loss: 3.4864e-06 - accuracy: 1.0000 - val_loss: 0.1552 - val_accuracy: 0.9821 Epoch 126/200 469/469 [==============================] - 2s 4ms/step - loss: 2.8008e-06 - accuracy: 1.0000 - val_loss: 0.1579 - val_accuracy: 0.9820 Epoch 127/200 469/469 [==============================] - 2s 3ms/step - loss: 2.2308e-06 - accuracy: 1.0000 - val_loss: 0.1603 - val_accuracy: 0.9820 Epoch 128/200 469/469 [==============================] - 2s 3ms/step - loss: 1.7629e-06 - accuracy: 1.0000 - val_loss: 0.1628 - val_accuracy: 0.9822 Epoch 129/200 469/469 [==============================] - 2s 4ms/step - loss: 1.4402e-06 - accuracy: 1.0000 - val_loss: 0.1647 - val_accuracy: 0.9822 Epoch 130/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1301e-06 - accuracy: 1.0000 - val_loss: 0.1672 - val_accuracy: 0.9822 Epoch 131/200 469/469 [==============================] - 2s 3ms/step - loss: 9.2444e-07 - accuracy: 1.0000 - val_loss: 0.1694 - val_accuracy: 0.9822 Epoch 132/200 469/469 [==============================] - 2s 3ms/step - loss: 7.4471e-07 - accuracy: 1.0000 - val_loss: 0.1718 - val_accuracy: 0.9824 Epoch 133/200 469/469 [==============================] - 2s 3ms/step - loss: 5.8486e-07 - accuracy: 1.0000 - val_loss: 0.1739 - val_accuracy: 0.9823 Epoch 134/200 469/469 [==============================] - 2s 3ms/step - loss: 4.6834e-07 - accuracy: 1.0000 - val_loss: 0.1761 - val_accuracy: 0.9822 Epoch 135/200 469/469 [==============================] - 2s 3ms/step - loss: 3.7385e-07 - accuracy: 1.0000 - val_loss: 0.1782 - val_accuracy: 0.9823 Epoch 136/200 469/469 [==============================] - 2s 4ms/step - loss: 3.1184e-07 - accuracy: 1.0000 - val_loss: 0.1801 - val_accuracy: 0.9824 Epoch 137/200 469/469 [==============================] - 2s 4ms/step - loss: 2.5207e-07 - accuracy: 1.0000 - val_loss: 0.1817 - val_accuracy: 0.9823 Epoch 138/200 469/469 [==============================] - 2s 3ms/step - loss: 1.9173e-07 - accuracy: 1.0000 - val_loss: 0.1836 - val_accuracy: 0.9823 Epoch 139/200 469/469 [==============================] - 2s 3ms/step - loss: 1.6292e-07 - accuracy: 1.0000 - val_loss: 0.1852 - val_accuracy: 0.9823 Epoch 140/200 469/469 [==============================] - 2s 3ms/step - loss: 1.3587e-07 - accuracy: 1.0000 - val_loss: 0.1870 - val_accuracy: 0.9823 Epoch 141/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1000e-07 - accuracy: 1.0000 - val_loss: 0.1889 - val_accuracy: 0.9823 Epoch 142/200 469/469 [==============================] - 2s 4ms/step - loss: 8.6909e-08 - accuracy: 1.0000 - val_loss: 0.1902 - val_accuracy: 0.9822 Epoch 143/200 469/469 [==============================] - 2s 3ms/step - loss: 6.9925e-08 - accuracy: 1.0000 - val_loss: 0.1916 - val_accuracy: 0.9822 Epoch 144/200 469/469 [==============================] - 2s 4ms/step - loss: 5.9532e-08 - accuracy: 1.0000 - val_loss: 0.1931 - val_accuracy: 0.9822 Epoch 145/200 469/469 [==============================] - 2s 3ms/step - loss: 5.0323e-08 - accuracy: 1.0000 - val_loss: 0.1942 - val_accuracy: 0.9822 Epoch 146/200 469/469 [==============================] - 2s 4ms/step - loss: 3.9739e-08 - accuracy: 1.0000 - val_loss: 0.1955 - val_accuracy: 0.9822 Epoch 147/200 469/469 [==============================] - 2s 4ms/step - loss: 2.8432e-08 - accuracy: 1.0000 - val_loss: 0.1966 - val_accuracy: 0.9823 Epoch 148/200 469/469 [==============================] - 2s 4ms/step - loss: 2.3689e-08 - accuracy: 1.0000 - val_loss: 0.1975 - val_accuracy: 0.9824 Epoch 149/200 469/469 [==============================] - 2s 3ms/step - loss: 2.0081e-08 - accuracy: 1.0000 - val_loss: 0.1987 - val_accuracy: 0.9824 Epoch 150/200 469/469 [==============================] - 2s 3ms/step - loss: 1.6603e-08 - accuracy: 1.0000 - val_loss: 0.1996 - val_accuracy: 0.9824 Epoch 151/200 469/469 [==============================] - 2s 3ms/step - loss: 1.4272e-08 - accuracy: 1.0000 - val_loss: 0.2004 - val_accuracy: 0.9824 Epoch 152/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2482e-08 - accuracy: 1.0000 - val_loss: 0.2012 - val_accuracy: 0.9824 Epoch 153/200 469/469 [==============================] - 2s 3ms/step - loss: 1.0863e-08 - accuracy: 1.0000 - val_loss: 0.2019 - val_accuracy: 0.9824 Epoch 154/200 469/469 [==============================] - 2s 3ms/step - loss: 9.5046e-09 - accuracy: 1.0000 - val_loss: 0.2026 - val_accuracy: 0.9824 Epoch 155/200 469/469 [==============================] - 2s 4ms/step - loss: 8.4040e-09 - accuracy: 1.0000 - val_loss: 0.2032 - val_accuracy: 0.9824 Epoch 156/200 469/469 [==============================] - 2s 3ms/step - loss: 7.7185e-09 - accuracy: 1.0000 - val_loss: 0.2038 - val_accuracy: 0.9824 Epoch 157/200 469/469 [==============================] - 2s 3ms/step - loss: 6.9795e-09 - accuracy: 1.0000 - val_loss: 0.2043 - val_accuracy: 0.9824 Epoch 158/200 469/469 [==============================] - 2s 3ms/step - loss: 6.1828e-09 - accuracy: 1.0000 - val_loss: 0.2048 - val_accuracy: 0.9824 Epoch 159/200 469/469 [==============================] - 2s 3ms/step - loss: 5.8014e-09 - accuracy: 1.0000 - val_loss: 0.2053 - val_accuracy: 0.9824 Epoch 160/200 469/469 [==============================] - 2s 3ms/step - loss: 5.3067e-09 - accuracy: 1.0000 - val_loss: 0.2058 - val_accuracy: 0.9825 Epoch 161/200 469/469 [==============================] - 2s 3ms/step - loss: 4.9471e-09 - accuracy: 1.0000 - val_loss: 0.2063 - val_accuracy: 0.9825 Epoch 162/200 469/469 [==============================] - 2s 3ms/step - loss: 4.6213e-09 - accuracy: 1.0000 - val_loss: 0.2067 - val_accuracy: 0.9825 Epoch 163/200 469/469 [==============================] - 2s 3ms/step - loss: 4.2776e-09 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9825 Epoch 164/200 469/469 [==============================] - 2s 3ms/step - loss: 4.0074e-09 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9825 Epoch 165/200 469/469 [==============================] - 2s 4ms/step - loss: 3.7570e-09 - accuracy: 1.0000 - val_loss: 0.2078 - val_accuracy: 0.9825 Epoch 166/200 469/469 [==============================] - 2s 4ms/step - loss: 3.5385e-09 - accuracy: 1.0000 - val_loss: 0.2082 - val_accuracy: 0.9825 Epoch 167/200 469/469 [==============================] - 2s 3ms/step - loss: 3.3656e-09 - accuracy: 1.0000 - val_loss: 0.2085 - val_accuracy: 0.9825 Epoch 168/200 469/469 [==============================] - 2s 3ms/step - loss: 3.2027e-09 - accuracy: 1.0000 - val_loss: 0.2088 - val_accuracy: 0.9825 Epoch 169/200 469/469 [==============================] - 2s 4ms/step - loss: 3.0497e-09 - accuracy: 1.0000 - val_loss: 0.2091 - val_accuracy: 0.9825 Epoch 170/200 469/469 [==============================] - 2s 4ms/step - loss: 2.8809e-09 - accuracy: 1.0000 - val_loss: 0.2094 - val_accuracy: 0.9825 Epoch 171/200 469/469 [==============================] - 2s 4ms/step - loss: 2.7557e-09 - accuracy: 1.0000 - val_loss: 0.2097 - val_accuracy: 0.9825 Epoch 172/200 469/469 [==============================] - 2s 4ms/step - loss: 2.6504e-09 - accuracy: 1.0000 - val_loss: 0.2099 - val_accuracy: 0.9825 Epoch 173/200 469/469 [==============================] - 2s 4ms/step - loss: 2.5411e-09 - accuracy: 1.0000 - val_loss: 0.2102 - val_accuracy: 0.9825 Epoch 174/200 469/469 [==============================] - 2s 4ms/step - loss: 2.4537e-09 - accuracy: 1.0000 - val_loss: 0.2105 - val_accuracy: 0.9826 Epoch 175/200 469/469 [==============================] - 2s 4ms/step - loss: 2.3345e-09 - accuracy: 1.0000 - val_loss: 0.2107 - val_accuracy: 0.9826 Epoch 176/200 469/469 [==============================] - 2s 4ms/step - loss: 2.2610e-09 - accuracy: 1.0000 - val_loss: 0.2110 - val_accuracy: 0.9827 Epoch 177/200 469/469 [==============================] - 2s 4ms/step - loss: 2.1716e-09 - accuracy: 1.0000 - val_loss: 0.2112 - val_accuracy: 0.9827 Epoch 178/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0722e-09 - accuracy: 1.0000 - val_loss: 0.2114 - val_accuracy: 0.9827 Epoch 179/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0206e-09 - accuracy: 1.0000 - val_loss: 0.2116 - val_accuracy: 0.9827 Epoch 180/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9431e-09 - accuracy: 1.0000 - val_loss: 0.2118 - val_accuracy: 0.9827 Epoch 181/200 469/469 [==============================] - 2s 4ms/step - loss: 1.8835e-09 - accuracy: 1.0000 - val_loss: 0.2120 - val_accuracy: 0.9827 Epoch 182/200 469/469 [==============================] - 2s 3ms/step - loss: 1.8438e-09 - accuracy: 1.0000 - val_loss: 0.2122 - val_accuracy: 0.9827 Epoch 183/200 469/469 [==============================] - 2s 4ms/step - loss: 1.7722e-09 - accuracy: 1.0000 - val_loss: 0.2124 - val_accuracy: 0.9827 Epoch 184/200 469/469 [==============================] - 2s 4ms/step - loss: 1.7385e-09 - accuracy: 1.0000 - val_loss: 0.2126 - val_accuracy: 0.9827 Epoch 185/200 469/469 [==============================] - 2s 4ms/step - loss: 1.6848e-09 - accuracy: 1.0000 - val_loss: 0.2128 - val_accuracy: 0.9827 Epoch 186/200 469/469 [==============================] - 2s 3ms/step - loss: 1.6510e-09 - accuracy: 1.0000 - val_loss: 0.2130 - val_accuracy: 0.9827 Epoch 187/200 469/469 [==============================] - 2s 4ms/step - loss: 1.6014e-09 - accuracy: 1.0000 - val_loss: 0.2131 - val_accuracy: 0.9827 Epoch 188/200 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-09 - accuracy: 1.0000 - val_loss: 0.2133 - val_accuracy: 0.9827 Epoch 189/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5199e-09 - accuracy: 1.0000 - val_loss: 0.2134 - val_accuracy: 0.9827 Epoch 190/200 469/469 [==============================] - 2s 3ms/step - loss: 1.4802e-09 - accuracy: 1.0000 - val_loss: 0.2136 - val_accuracy: 0.9827 Epoch 191/200 469/469 [==============================] - 2s 4ms/step - loss: 1.4424e-09 - accuracy: 1.0000 - val_loss: 0.2138 - val_accuracy: 0.9827 Epoch 192/200 469/469 [==============================] - 2s 3ms/step - loss: 1.4047e-09 - accuracy: 1.0000 - val_loss: 0.2139 - val_accuracy: 0.9828 Epoch 193/200 469/469 [==============================] - 2s 4ms/step - loss: 1.3709e-09 - accuracy: 1.0000 - val_loss: 0.2141 - val_accuracy: 0.9828 Epoch 194/200 469/469 [==============================] - 2s 4ms/step - loss: 1.3252e-09 - accuracy: 1.0000 - val_loss: 0.2142 - val_accuracy: 0.9828 Epoch 195/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2974e-09 - accuracy: 1.0000 - val_loss: 0.2143 - val_accuracy: 0.9828 Epoch 196/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2835e-09 - accuracy: 1.0000 - val_loss: 0.2145 - val_accuracy: 0.9828 Epoch 197/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2418e-09 - accuracy: 1.0000 - val_loss: 0.2146 - val_accuracy: 0.9828 Epoch 198/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2239e-09 - accuracy: 1.0000 - val_loss: 0.2147 - val_accuracy: 0.9828 Epoch 199/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2080e-09 - accuracy: 1.0000 - val_loss: 0.2148 - val_accuracy: 0.9828 Epoch 200/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1722e-09 - accuracy: 1.0000 - val_loss: 0.2150 - val_accuracy: 0.9828
(b) Activation function: the logistic sigmoid function; initialization: Xavier initializer¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.GlorotNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
#tf.keras.layers.Dense(512,kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_4" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_4 (Flatten) (None, 784) 0 _________________________________________________________________ dense_24 (Dense) (None, 512) 401920 _________________________________________________________________ dense_25 (Dense) (None, 512) 262656 _________________________________________________________________ dense_26 (Dense) (None, 512) 262656 _________________________________________________________________ dense_27 (Dense) (None, 512) 262656 _________________________________________________________________ dense_28 (Dense) (None, 512) 262656 _________________________________________________________________ dense_29 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam() #learning_rate=1.0 for SGD
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
historya1 = model.fit(X_train, y_train, batch_size=128, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 469/469 [==============================] - 2s 4ms/step - loss: 0.8131 - accuracy: 0.7149 - val_loss: 0.3044 - val_accuracy: 0.9094 Epoch 2/200 469/469 [==============================] - 2s 3ms/step - loss: 0.2390 - accuracy: 0.9298 - val_loss: 0.1955 - val_accuracy: 0.9406 Epoch 3/200 469/469 [==============================] - 2s 3ms/step - loss: 0.1628 - accuracy: 0.9516 - val_loss: 0.1500 - val_accuracy: 0.9555 Epoch 4/200 469/469 [==============================] - 2s 4ms/step - loss: 0.1233 - accuracy: 0.9635 - val_loss: 0.1188 - val_accuracy: 0.9632 Epoch 5/200 469/469 [==============================] - 2s 4ms/step - loss: 0.1013 - accuracy: 0.9692 - val_loss: 0.1172 - val_accuracy: 0.9663 Epoch 6/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0859 - accuracy: 0.9740 - val_loss: 0.1048 - val_accuracy: 0.9698 Epoch 7/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0694 - accuracy: 0.9795 - val_loss: 0.0946 - val_accuracy: 0.9743 Epoch 8/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0593 - accuracy: 0.9819 - val_loss: 0.0973 - val_accuracy: 0.9715 Epoch 9/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0506 - accuracy: 0.9850 - val_loss: 0.0898 - val_accuracy: 0.9745 Epoch 10/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0453 - accuracy: 0.9863 - val_loss: 0.0797 - val_accuracy: 0.9783 Epoch 11/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0381 - accuracy: 0.9887 - val_loss: 0.0824 - val_accuracy: 0.9777 Epoch 12/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0323 - accuracy: 0.9905 - val_loss: 0.0952 - val_accuracy: 0.9770 Epoch 13/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0314 - accuracy: 0.9908 - val_loss: 0.0864 - val_accuracy: 0.9777 Epoch 14/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0251 - accuracy: 0.9919 - val_loss: 0.0930 - val_accuracy: 0.9754 Epoch 15/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0255 - accuracy: 0.9921 - val_loss: 0.0851 - val_accuracy: 0.9795 Epoch 16/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0201 - accuracy: 0.9937 - val_loss: 0.0877 - val_accuracy: 0.9782 Epoch 17/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0179 - accuracy: 0.9946 - val_loss: 0.0958 - val_accuracy: 0.9803 Epoch 18/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0173 - accuracy: 0.9947 - val_loss: 0.0897 - val_accuracy: 0.9795 Epoch 19/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0163 - accuracy: 0.9952 - val_loss: 0.0908 - val_accuracy: 0.9793 Epoch 20/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0151 - accuracy: 0.9952 - val_loss: 0.1051 - val_accuracy: 0.9779 Epoch 21/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0159 - accuracy: 0.9951 - val_loss: 0.0798 - val_accuracy: 0.9816 Epoch 22/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0118 - accuracy: 0.9967 - val_loss: 0.0909 - val_accuracy: 0.9789 Epoch 23/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0111 - accuracy: 0.9966 - val_loss: 0.1012 - val_accuracy: 0.9795 Epoch 24/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9963 - val_loss: 0.0869 - val_accuracy: 0.9827 Epoch 25/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0106 - accuracy: 0.9969 - val_loss: 0.1158 - val_accuracy: 0.9773 Epoch 26/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9972 - val_loss: 0.1081 - val_accuracy: 0.9808 Epoch 27/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0099 - accuracy: 0.9971 - val_loss: 0.0969 - val_accuracy: 0.9809 Epoch 28/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0096 - accuracy: 0.9968 - val_loss: 0.1194 - val_accuracy: 0.9783 Epoch 29/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0080 - accuracy: 0.9975 - val_loss: 0.0938 - val_accuracy: 0.9831 Epoch 30/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0069 - accuracy: 0.9981 - val_loss: 0.1113 - val_accuracy: 0.9808 Epoch 31/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0080 - accuracy: 0.9976 - val_loss: 0.1206 - val_accuracy: 0.9790 Epoch 32/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0068 - accuracy: 0.9980 - val_loss: 0.1048 - val_accuracy: 0.9812 Epoch 33/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0076 - accuracy: 0.9977 - val_loss: 0.1038 - val_accuracy: 0.9824 Epoch 34/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0073 - accuracy: 0.9978 - val_loss: 0.0916 - val_accuracy: 0.9829 Epoch 35/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0047 - accuracy: 0.9987 - val_loss: 0.0897 - val_accuracy: 0.9834 Epoch 36/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0072 - accuracy: 0.9977 - val_loss: 0.0945 - val_accuracy: 0.9819 Epoch 37/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0056 - accuracy: 0.9982 - val_loss: 0.1134 - val_accuracy: 0.9816 Epoch 38/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0056 - accuracy: 0.9984 - val_loss: 0.1114 - val_accuracy: 0.9818 Epoch 39/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0066 - accuracy: 0.9979 - val_loss: 0.1002 - val_accuracy: 0.9820 Epoch 40/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0064 - accuracy: 0.9981 - val_loss: 0.1066 - val_accuracy: 0.9802 Epoch 41/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0043 - accuracy: 0.9988 - val_loss: 0.1112 - val_accuracy: 0.9803 Epoch 42/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0043 - accuracy: 0.9986 - val_loss: 0.1058 - val_accuracy: 0.9829 Epoch 43/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0038 - accuracy: 0.9989 - val_loss: 0.1122 - val_accuracy: 0.9833 Epoch 44/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0039 - accuracy: 0.9988 - val_loss: 0.1446 - val_accuracy: 0.9802 Epoch 45/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9983 - val_loss: 0.1108 - val_accuracy: 0.9818 Epoch 46/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0023 - accuracy: 0.9992 - val_loss: 0.1388 - val_accuracy: 0.9789 Epoch 47/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0063 - accuracy: 0.9984 - val_loss: 0.1019 - val_accuracy: 0.9818 Epoch 48/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0039 - accuracy: 0.9988 - val_loss: 0.1167 - val_accuracy: 0.9830 Epoch 49/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0026 - accuracy: 0.9991 - val_loss: 0.1184 - val_accuracy: 0.9821 Epoch 50/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0048 - accuracy: 0.9988 - val_loss: 0.1055 - val_accuracy: 0.9842 Epoch 51/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0046 - accuracy: 0.9987 - val_loss: 0.1385 - val_accuracy: 0.9770 Epoch 52/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0044 - accuracy: 0.9988 - val_loss: 0.1125 - val_accuracy: 0.9823 Epoch 53/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0024 - accuracy: 0.9993 - val_loss: 0.1224 - val_accuracy: 0.9837 Epoch 54/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0062 - accuracy: 0.9984 - val_loss: 0.1171 - val_accuracy: 0.9831 Epoch 55/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0045 - accuracy: 0.9987 - val_loss: 0.1130 - val_accuracy: 0.9830 Epoch 56/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0040 - accuracy: 0.9990 - val_loss: 0.1011 - val_accuracy: 0.9838 Epoch 57/200 469/469 [==============================] - 2s 4ms/step - loss: 9.1745e-04 - accuracy: 0.9997 - val_loss: 0.1178 - val_accuracy: 0.9837 Epoch 58/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0026 - accuracy: 0.9992 - val_loss: 0.1208 - val_accuracy: 0.9815 Epoch 59/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0058 - accuracy: 0.9985 - val_loss: 0.0981 - val_accuracy: 0.9837 Epoch 60/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0027 - accuracy: 0.9991 - val_loss: 0.1092 - val_accuracy: 0.9845 Epoch 61/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0044 - accuracy: 0.9988 - val_loss: 0.1207 - val_accuracy: 0.9823 Epoch 62/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0011 - accuracy: 0.9998 - val_loss: 0.1155 - val_accuracy: 0.9844 Epoch 63/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2653e-04 - accuracy: 1.0000 - val_loss: 0.1228 - val_accuracy: 0.9843 Epoch 64/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0071 - accuracy: 0.9982 - val_loss: 0.1367 - val_accuracy: 0.9791 Epoch 65/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0040 - accuracy: 0.9990 - val_loss: 0.1177 - val_accuracy: 0.9829 Epoch 66/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0038 - accuracy: 0.9990 - val_loss: 0.1024 - val_accuracy: 0.9844 Epoch 67/200 469/469 [==============================] - 2s 3ms/step - loss: 4.1127e-04 - accuracy: 0.9999 - val_loss: 0.1101 - val_accuracy: 0.9847 Epoch 68/200 469/469 [==============================] - 2s 3ms/step - loss: 5.6425e-05 - accuracy: 1.0000 - val_loss: 0.1149 - val_accuracy: 0.9852 Epoch 69/200 469/469 [==============================] - 2s 3ms/step - loss: 2.1659e-05 - accuracy: 1.0000 - val_loss: 0.1184 - val_accuracy: 0.9855 Epoch 70/200 469/469 [==============================] - 2s 3ms/step - loss: 1.0396e-05 - accuracy: 1.0000 - val_loss: 0.1226 - val_accuracy: 0.9854 Epoch 71/200 469/469 [==============================] - 2s 4ms/step - loss: 7.1113e-06 - accuracy: 1.0000 - val_loss: 0.1263 - val_accuracy: 0.9854 Epoch 72/200 469/469 [==============================] - 2s 3ms/step - loss: 4.9408e-06 - accuracy: 1.0000 - val_loss: 0.1299 - val_accuracy: 0.9852 Epoch 73/200 469/469 [==============================] - 2s 4ms/step - loss: 3.5696e-06 - accuracy: 1.0000 - val_loss: 0.1331 - val_accuracy: 0.9851 Epoch 74/200 469/469 [==============================] - 2s 4ms/step - loss: 2.6280e-06 - accuracy: 1.0000 - val_loss: 0.1361 - val_accuracy: 0.9851 Epoch 75/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0153e-06 - accuracy: 1.0000 - val_loss: 0.1391 - val_accuracy: 0.9852 Epoch 76/200 469/469 [==============================] - 2s 4ms/step - loss: 1.5774e-06 - accuracy: 1.0000 - val_loss: 0.1419 - val_accuracy: 0.9850 Epoch 77/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2338e-06 - accuracy: 1.0000 - val_loss: 0.1446 - val_accuracy: 0.9849 Epoch 78/200 469/469 [==============================] - 2s 3ms/step - loss: 9.7848e-07 - accuracy: 1.0000 - val_loss: 0.1473 - val_accuracy: 0.9849 Epoch 79/200 469/469 [==============================] - 2s 3ms/step - loss: 7.7092e-07 - accuracy: 1.0000 - val_loss: 0.1499 - val_accuracy: 0.9849 Epoch 80/200 469/469 [==============================] - 2s 4ms/step - loss: 6.2419e-07 - accuracy: 1.0000 - val_loss: 0.1523 - val_accuracy: 0.9850 Epoch 81/200 469/469 [==============================] - 2s 3ms/step - loss: 4.9624e-07 - accuracy: 1.0000 - val_loss: 0.1549 - val_accuracy: 0.9849 Epoch 82/200 469/469 [==============================] - 2s 4ms/step - loss: 3.9337e-07 - accuracy: 1.0000 - val_loss: 0.1571 - val_accuracy: 0.9849 Epoch 83/200 469/469 [==============================] - 2s 3ms/step - loss: 3.1382e-07 - accuracy: 1.0000 - val_loss: 0.1595 - val_accuracy: 0.9849 Epoch 84/200 469/469 [==============================] - 2s 3ms/step - loss: 2.4437e-07 - accuracy: 1.0000 - val_loss: 0.1616 - val_accuracy: 0.9849 Epoch 85/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9419e-07 - accuracy: 1.0000 - val_loss: 0.1637 - val_accuracy: 0.9849 Epoch 86/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5399e-07 - accuracy: 1.0000 - val_loss: 0.1656 - val_accuracy: 0.9850 Epoch 87/200 469/469 [==============================] - 2s 4ms/step - loss: 1.2316e-07 - accuracy: 1.0000 - val_loss: 0.1674 - val_accuracy: 0.9851 Epoch 88/200 469/469 [==============================] - 2s 3ms/step - loss: 9.9206e-08 - accuracy: 1.0000 - val_loss: 0.1691 - val_accuracy: 0.9852 Epoch 89/200 469/469 [==============================] - 2s 3ms/step - loss: 8.0772e-08 - accuracy: 1.0000 - val_loss: 0.1708 - val_accuracy: 0.9853 Epoch 90/200 469/469 [==============================] - 2s 3ms/step - loss: 6.6904e-08 - accuracy: 1.0000 - val_loss: 0.1725 - val_accuracy: 0.9853 Epoch 91/200 469/469 [==============================] - 2s 3ms/step - loss: 5.5542e-08 - accuracy: 1.0000 - val_loss: 0.1741 - val_accuracy: 0.9853 Epoch 92/200 469/469 [==============================] - 2s 3ms/step - loss: 4.5446e-08 - accuracy: 1.0000 - val_loss: 0.1756 - val_accuracy: 0.9852 Epoch 93/200 469/469 [==============================] - 2s 4ms/step - loss: 3.8587e-08 - accuracy: 1.0000 - val_loss: 0.1770 - val_accuracy: 0.9853 Epoch 94/200 469/469 [==============================] - 2s 4ms/step - loss: 3.2603e-08 - accuracy: 1.0000 - val_loss: 0.1784 - val_accuracy: 0.9853 Epoch 95/200 469/469 [==============================] - 2s 4ms/step - loss: 2.7673e-08 - accuracy: 1.0000 - val_loss: 0.1797 - val_accuracy: 0.9853 Epoch 96/200 469/469 [==============================] - 2s 4ms/step - loss: 2.3509e-08 - accuracy: 1.0000 - val_loss: 0.1810 - val_accuracy: 0.9853 Epoch 97/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0257e-08 - accuracy: 1.0000 - val_loss: 0.1821 - val_accuracy: 0.9853 Epoch 98/200 469/469 [==============================] - 2s 3ms/step - loss: 1.7350e-08 - accuracy: 1.0000 - val_loss: 0.1833 - val_accuracy: 0.9853 Epoch 99/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5109e-08 - accuracy: 1.0000 - val_loss: 0.1843 - val_accuracy: 0.9853 Epoch 100/200 469/469 [==============================] - 2s 3ms/step - loss: 1.3103e-08 - accuracy: 1.0000 - val_loss: 0.1854 - val_accuracy: 0.9854 Epoch 101/200 469/469 [==============================] - 2s 4ms/step - loss: 1.1583e-08 - accuracy: 1.0000 - val_loss: 0.1863 - val_accuracy: 0.9856 Epoch 102/200 469/469 [==============================] - 2s 4ms/step - loss: 1.0198e-08 - accuracy: 1.0000 - val_loss: 0.1872 - val_accuracy: 0.9856 Epoch 103/200 469/469 [==============================] - 2s 4ms/step - loss: 8.9962e-09 - accuracy: 1.0000 - val_loss: 0.1882 - val_accuracy: 0.9854 Epoch 104/200 469/469 [==============================] - 2s 4ms/step - loss: 8.0803e-09 - accuracy: 1.0000 - val_loss: 0.1890 - val_accuracy: 0.9854 Epoch 105/200 469/469 [==============================] - 2s 4ms/step - loss: 7.2180e-09 - accuracy: 1.0000 - val_loss: 0.1899 - val_accuracy: 0.9855 Epoch 106/200 469/469 [==============================] - 2s 4ms/step - loss: 6.5008e-09 - accuracy: 1.0000 - val_loss: 0.1906 - val_accuracy: 0.9856 Epoch 107/200 469/469 [==============================] - 2s 4ms/step - loss: 5.9346e-09 - accuracy: 1.0000 - val_loss: 0.1914 - val_accuracy: 0.9856 Epoch 108/200 469/469 [==============================] - 2s 4ms/step - loss: 5.3544e-09 - accuracy: 1.0000 - val_loss: 0.1920 - val_accuracy: 0.9856 Epoch 109/200 469/469 [==============================] - 2s 3ms/step - loss: 4.9054e-09 - accuracy: 1.0000 - val_loss: 0.1926 - val_accuracy: 0.9856 Epoch 110/200 469/469 [==============================] - 2s 3ms/step - loss: 4.4783e-09 - accuracy: 1.0000 - val_loss: 0.1932 - val_accuracy: 0.9856 Epoch 111/200 469/469 [==============================] - 2s 3ms/step - loss: 4.1524e-09 - accuracy: 1.0000 - val_loss: 0.1937 - val_accuracy: 0.9856 Epoch 112/200 469/469 [==============================] - 2s 3ms/step - loss: 3.8842e-09 - accuracy: 1.0000 - val_loss: 0.1943 - val_accuracy: 0.9855 Epoch 113/200 469/469 [==============================] - 2s 3ms/step - loss: 3.6080e-09 - accuracy: 1.0000 - val_loss: 0.1948 - val_accuracy: 0.9855 Epoch 114/200 469/469 [==============================] - 2s 4ms/step - loss: 3.3418e-09 - accuracy: 1.0000 - val_loss: 0.1952 - val_accuracy: 0.9855 Epoch 115/200 469/469 [==============================] - 2s 4ms/step - loss: 3.1094e-09 - accuracy: 1.0000 - val_loss: 0.1957 - val_accuracy: 0.9855 Epoch 116/200 469/469 [==============================] - 2s 4ms/step - loss: 2.9445e-09 - accuracy: 1.0000 - val_loss: 0.1961 - val_accuracy: 0.9854 Epoch 117/200 469/469 [==============================] - 2s 4ms/step - loss: 2.7756e-09 - accuracy: 1.0000 - val_loss: 0.1965 - val_accuracy: 0.9854 Epoch 118/200 469/469 [==============================] - 2s 4ms/step - loss: 2.6425e-09 - accuracy: 1.0000 - val_loss: 0.1969 - val_accuracy: 0.9853 Epoch 119/200 469/469 [==============================] - 2s 4ms/step - loss: 2.5074e-09 - accuracy: 1.0000 - val_loss: 0.1973 - val_accuracy: 0.9853 Epoch 120/200 469/469 [==============================] - 2s 4ms/step - loss: 2.3583e-09 - accuracy: 1.0000 - val_loss: 0.1976 - val_accuracy: 0.9853 Epoch 121/200 469/469 [==============================] - 2s 4ms/step - loss: 2.2471e-09 - accuracy: 1.0000 - val_loss: 0.1979 - val_accuracy: 0.9853 Epoch 122/200 469/469 [==============================] - 2s 4ms/step - loss: 2.1696e-09 - accuracy: 1.0000 - val_loss: 0.1982 - val_accuracy: 0.9853 Epoch 123/200 469/469 [==============================] - 2s 3ms/step - loss: 2.0643e-09 - accuracy: 1.0000 - val_loss: 0.1985 - val_accuracy: 0.9853 Epoch 124/200 469/469 [==============================] - 2s 3ms/step - loss: 1.9888e-09 - accuracy: 1.0000 - val_loss: 0.1988 - val_accuracy: 0.9853 Epoch 125/200 469/469 [==============================] - 2s 3ms/step - loss: 1.9073e-09 - accuracy: 1.0000 - val_loss: 0.1991 - val_accuracy: 0.9853 Epoch 126/200 469/469 [==============================] - 2s 3ms/step - loss: 1.8299e-09 - accuracy: 1.0000 - val_loss: 0.1994 - val_accuracy: 0.9853 Epoch 127/200 469/469 [==============================] - 2s 3ms/step - loss: 1.7563e-09 - accuracy: 1.0000 - val_loss: 0.1996 - val_accuracy: 0.9853 Epoch 128/200 469/469 [==============================] - 2s 3ms/step - loss: 1.7027e-09 - accuracy: 1.0000 - val_loss: 0.1999 - val_accuracy: 0.9853 Epoch 129/200 469/469 [==============================] - 2s 3ms/step - loss: 1.6491e-09 - accuracy: 1.0000 - val_loss: 0.2001 - val_accuracy: 0.9852 Epoch 130/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5875e-09 - accuracy: 1.0000 - val_loss: 0.2004 - val_accuracy: 0.9852 Epoch 131/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5517e-09 - accuracy: 1.0000 - val_loss: 0.2006 - val_accuracy: 0.9852 Epoch 132/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5100e-09 - accuracy: 1.0000 - val_loss: 0.2009 - val_accuracy: 0.9852 Epoch 133/200 469/469 [==============================] - 2s 4ms/step - loss: 1.4603e-09 - accuracy: 1.0000 - val_loss: 0.2011 - val_accuracy: 0.9852 Epoch 134/200 469/469 [==============================] - 2s 4ms/step - loss: 1.4226e-09 - accuracy: 1.0000 - val_loss: 0.2013 - val_accuracy: 0.9852 Epoch 135/200 469/469 [==============================] - 2s 3ms/step - loss: 1.3669e-09 - accuracy: 1.0000 - val_loss: 0.2015 - val_accuracy: 0.9852 Epoch 136/200 469/469 [==============================] - 2s 3ms/step - loss: 1.3351e-09 - accuracy: 1.0000 - val_loss: 0.2017 - val_accuracy: 0.9852 Epoch 137/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2994e-09 - accuracy: 1.0000 - val_loss: 0.2019 - val_accuracy: 0.9852 Epoch 138/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2557e-09 - accuracy: 1.0000 - val_loss: 0.2021 - val_accuracy: 0.9851 Epoch 139/200 469/469 [==============================] - 2s 4ms/step - loss: 1.2279e-09 - accuracy: 1.0000 - val_loss: 0.2023 - val_accuracy: 0.9851 Epoch 140/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1901e-09 - accuracy: 1.0000 - val_loss: 0.2025 - val_accuracy: 0.9851 Epoch 141/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1702e-09 - accuracy: 1.0000 - val_loss: 0.2026 - val_accuracy: 0.9851 Epoch 142/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1265e-09 - accuracy: 1.0000 - val_loss: 0.2028 - val_accuracy: 0.9851 Epoch 143/200 469/469 [==============================] - 2s 3ms/step - loss: 1.0908e-09 - accuracy: 1.0000 - val_loss: 0.2030 - val_accuracy: 0.9851 Epoch 144/200 469/469 [==============================] - 2s 3ms/step - loss: 1.0729e-09 - accuracy: 1.0000 - val_loss: 0.2031 - val_accuracy: 0.9851 Epoch 145/200 469/469 [==============================] - 2s 3ms/step - loss: 1.0471e-09 - accuracy: 1.0000 - val_loss: 0.2033 - val_accuracy: 0.9851 Epoch 146/200 469/469 [==============================] - 2s 3ms/step - loss: 1.0212e-09 - accuracy: 1.0000 - val_loss: 0.2034 - val_accuracy: 0.9851 Epoch 147/200 469/469 [==============================] - 2s 3ms/step - loss: 9.8546e-10 - accuracy: 1.0000 - val_loss: 0.2036 - val_accuracy: 0.9851 Epoch 148/200 469/469 [==============================] - 2s 3ms/step - loss: 9.6758e-10 - accuracy: 1.0000 - val_loss: 0.2037 - val_accuracy: 0.9851 Epoch 149/200 469/469 [==============================] - 2s 3ms/step - loss: 9.5367e-10 - accuracy: 1.0000 - val_loss: 0.2038 - val_accuracy: 0.9851 Epoch 150/200 469/469 [==============================] - 2s 3ms/step - loss: 9.2983e-10 - accuracy: 1.0000 - val_loss: 0.2040 - val_accuracy: 0.9851 Epoch 151/200 469/469 [==============================] - 2s 4ms/step - loss: 9.1394e-10 - accuracy: 1.0000 - val_loss: 0.2041 - val_accuracy: 0.9851 Epoch 152/200 469/469 [==============================] - 2s 3ms/step - loss: 9.0202e-10 - accuracy: 1.0000 - val_loss: 0.2043 - val_accuracy: 0.9851 Epoch 153/200 469/469 [==============================] - 2s 4ms/step - loss: 8.8811e-10 - accuracy: 1.0000 - val_loss: 0.2044 - val_accuracy: 0.9851 Epoch 154/200 469/469 [==============================] - 2s 3ms/step - loss: 8.7023e-10 - accuracy: 1.0000 - val_loss: 0.2045 - val_accuracy: 0.9851 Epoch 155/200 469/469 [==============================] - 2s 3ms/step - loss: 8.5632e-10 - accuracy: 1.0000 - val_loss: 0.2047 - val_accuracy: 0.9851 Epoch 156/200 469/469 [==============================] - 2s 4ms/step - loss: 8.3446e-10 - accuracy: 1.0000 - val_loss: 0.2048 - val_accuracy: 0.9851 Epoch 157/200 469/469 [==============================] - 2s 3ms/step - loss: 8.1460e-10 - accuracy: 1.0000 - val_loss: 0.2049 - val_accuracy: 0.9851 Epoch 158/200 469/469 [==============================] - 2s 4ms/step - loss: 8.0665e-10 - accuracy: 1.0000 - val_loss: 0.2050 - val_accuracy: 0.9851 Epoch 159/200 469/469 [==============================] - 2s 3ms/step - loss: 7.9075e-10 - accuracy: 1.0000 - val_loss: 0.2051 - val_accuracy: 0.9851 Epoch 160/200 469/469 [==============================] - 2s 3ms/step - loss: 7.8479e-10 - accuracy: 1.0000 - val_loss: 0.2053 - val_accuracy: 0.9851 Epoch 161/200 469/469 [==============================] - 2s 3ms/step - loss: 7.6691e-10 - accuracy: 1.0000 - val_loss: 0.2054 - val_accuracy: 0.9851 Epoch 162/200 469/469 [==============================] - 2s 3ms/step - loss: 7.5102e-10 - accuracy: 1.0000 - val_loss: 0.2055 - val_accuracy: 0.9851 Epoch 163/200 469/469 [==============================] - 2s 4ms/step - loss: 7.3711e-10 - accuracy: 1.0000 - val_loss: 0.2056 - val_accuracy: 0.9851 Epoch 164/200 469/469 [==============================] - 2s 4ms/step - loss: 7.2916e-10 - accuracy: 1.0000 - val_loss: 0.2057 - val_accuracy: 0.9851 Epoch 165/200 469/469 [==============================] - 2s 4ms/step - loss: 7.0929e-10 - accuracy: 1.0000 - val_loss: 0.2058 - val_accuracy: 0.9851 Epoch 166/200 469/469 [==============================] - 2s 3ms/step - loss: 6.9141e-10 - accuracy: 1.0000 - val_loss: 0.2059 - val_accuracy: 0.9851 Epoch 167/200 469/469 [==============================] - 2s 4ms/step - loss: 6.8347e-10 - accuracy: 1.0000 - val_loss: 0.2060 - val_accuracy: 0.9851 Epoch 168/200 469/469 [==============================] - 2s 4ms/step - loss: 6.6956e-10 - accuracy: 1.0000 - val_loss: 0.2061 - val_accuracy: 0.9851 Epoch 169/200 469/469 [==============================] - 2s 4ms/step - loss: 6.6161e-10 - accuracy: 1.0000 - val_loss: 0.2062 - val_accuracy: 0.9851 Epoch 170/200 469/469 [==============================] - 2s 4ms/step - loss: 6.4969e-10 - accuracy: 1.0000 - val_loss: 0.2063 - val_accuracy: 0.9851 Epoch 171/200 469/469 [==============================] - 2s 4ms/step - loss: 6.3380e-10 - accuracy: 1.0000 - val_loss: 0.2064 - val_accuracy: 0.9851 Epoch 172/200 469/469 [==============================] - 2s 3ms/step - loss: 6.2585e-10 - accuracy: 1.0000 - val_loss: 0.2065 - val_accuracy: 0.9851 Epoch 173/200 469/469 [==============================] - 2s 3ms/step - loss: 6.1591e-10 - accuracy: 1.0000 - val_loss: 0.2066 - val_accuracy: 0.9851 Epoch 174/200 469/469 [==============================] - 2s 4ms/step - loss: 6.0797e-10 - accuracy: 1.0000 - val_loss: 0.2066 - val_accuracy: 0.9851 Epoch 175/200 469/469 [==============================] - 2s 4ms/step - loss: 6.0399e-10 - accuracy: 1.0000 - val_loss: 0.2067 - val_accuracy: 0.9851 Epoch 176/200 469/469 [==============================] - 2s 3ms/step - loss: 5.8611e-10 - accuracy: 1.0000 - val_loss: 0.2068 - val_accuracy: 0.9851 Epoch 177/200 469/469 [==============================] - 2s 3ms/step - loss: 5.8214e-10 - accuracy: 1.0000 - val_loss: 0.2069 - val_accuracy: 0.9851 Epoch 178/200 469/469 [==============================] - 2s 3ms/step - loss: 5.7022e-10 - accuracy: 1.0000 - val_loss: 0.2070 - val_accuracy: 0.9851 Epoch 179/200 469/469 [==============================] - 2s 3ms/step - loss: 5.7022e-10 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9851 Epoch 180/200 469/469 [==============================] - 2s 3ms/step - loss: 5.6426e-10 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9851 Epoch 181/200 469/469 [==============================] - 2s 3ms/step - loss: 5.5035e-10 - accuracy: 1.0000 - val_loss: 0.2072 - val_accuracy: 0.9851 Epoch 182/200 469/469 [==============================] - 2s 3ms/step - loss: 5.4240e-10 - accuracy: 1.0000 - val_loss: 0.2073 - val_accuracy: 0.9851 Epoch 183/200 469/469 [==============================] - 2s 3ms/step - loss: 5.3644e-10 - accuracy: 1.0000 - val_loss: 0.2074 - val_accuracy: 0.9851 Epoch 184/200 469/469 [==============================] - 2s 4ms/step - loss: 5.3247e-10 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9851 Epoch 185/200 469/469 [==============================] - 2s 3ms/step - loss: 5.2253e-10 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9851 Epoch 186/200 469/469 [==============================] - 2s 3ms/step - loss: 5.1657e-10 - accuracy: 1.0000 - val_loss: 0.2076 - val_accuracy: 0.9851 Epoch 187/200 469/469 [==============================] - 2s 4ms/step - loss: 5.0863e-10 - accuracy: 1.0000 - val_loss: 0.2077 - val_accuracy: 0.9851 Epoch 188/200 469/469 [==============================] - 2s 3ms/step - loss: 5.0068e-10 - accuracy: 1.0000 - val_loss: 0.2077 - val_accuracy: 0.9851 Epoch 189/200 469/469 [==============================] - 2s 4ms/step - loss: 4.9670e-10 - accuracy: 1.0000 - val_loss: 0.2078 - val_accuracy: 0.9851 Epoch 190/200 469/469 [==============================] - 2s 4ms/step - loss: 4.9273e-10 - accuracy: 1.0000 - val_loss: 0.2079 - val_accuracy: 0.9851 Epoch 191/200 469/469 [==============================] - 2s 3ms/step - loss: 4.8478e-10 - accuracy: 1.0000 - val_loss: 0.2080 - val_accuracy: 0.9851 Epoch 192/200 469/469 [==============================] - 2s 3ms/step - loss: 4.8081e-10 - accuracy: 1.0000 - val_loss: 0.2080 - val_accuracy: 0.9851 Epoch 193/200 469/469 [==============================] - 2s 3ms/step - loss: 4.6889e-10 - accuracy: 1.0000 - val_loss: 0.2081 - val_accuracy: 0.9851 Epoch 194/200 469/469 [==============================] - 2s 4ms/step - loss: 4.6293e-10 - accuracy: 1.0000 - val_loss: 0.2081 - val_accuracy: 0.9851 Epoch 195/200 469/469 [==============================] - 2s 4ms/step - loss: 4.5299e-10 - accuracy: 1.0000 - val_loss: 0.2082 - val_accuracy: 0.9851 Epoch 196/200 469/469 [==============================] - 2s 3ms/step - loss: 4.4902e-10 - accuracy: 1.0000 - val_loss: 0.2083 - val_accuracy: 0.9851 Epoch 197/200 469/469 [==============================] - 2s 4ms/step - loss: 4.4306e-10 - accuracy: 1.0000 - val_loss: 0.2083 - val_accuracy: 0.9852 Epoch 198/200 469/469 [==============================] - 2s 4ms/step - loss: 4.4306e-10 - accuracy: 1.0000 - val_loss: 0.2084 - val_accuracy: 0.9852 Epoch 199/200 469/469 [==============================] - 2s 4ms/step - loss: 4.3313e-10 - accuracy: 1.0000 - val_loss: 0.2084 - val_accuracy: 0.9852 Epoch 200/200 469/469 [==============================] - 2s 4ms/step - loss: 4.3114e-10 - accuracy: 1.0000 - val_loss: 0.2085 - val_accuracy: 0.9852
(c) Activation function: ReLU; initialization: random numbers generated from the normal¶
distribution (μ = 0, σ = 0.01)
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.01, seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
#tf.keras.layers.Dense(512,kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_5" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_5 (Flatten) (None, 784) 0 _________________________________________________________________ dense_30 (Dense) (None, 512) 401920 _________________________________________________________________ dense_31 (Dense) (None, 512) 262656 _________________________________________________________________ dense_32 (Dense) (None, 512) 262656 _________________________________________________________________ dense_33 (Dense) (None, 512) 262656 _________________________________________________________________ dense_34 (Dense) (None, 512) 262656 _________________________________________________________________ dense_35 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam() #learning_rate=1.0 for SGD
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
historya2 = model.fit(X_train, y_train, batch_size=128, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 469/469 [==============================] - 2s 4ms/step - loss: 0.5278 - accuracy: 0.8253 - val_loss: 0.1801 - val_accuracy: 0.9498 Epoch 2/200 469/469 [==============================] - 2s 4ms/step - loss: 0.1554 - accuracy: 0.9572 - val_loss: 0.1551 - val_accuracy: 0.9558 Epoch 3/200 469/469 [==============================] - 2s 3ms/step - loss: 0.1022 - accuracy: 0.9715 - val_loss: 0.1108 - val_accuracy: 0.9672 Epoch 4/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0734 - accuracy: 0.9793 - val_loss: 0.0868 - val_accuracy: 0.9764 Epoch 5/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0590 - accuracy: 0.9827 - val_loss: 0.0990 - val_accuracy: 0.9731 Epoch 6/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0470 - accuracy: 0.9862 - val_loss: 0.1011 - val_accuracy: 0.9734 Epoch 7/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0438 - accuracy: 0.9873 - val_loss: 0.0937 - val_accuracy: 0.9772 Epoch 8/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0351 - accuracy: 0.9903 - val_loss: 0.0861 - val_accuracy: 0.9793 Epoch 9/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0320 - accuracy: 0.9905 - val_loss: 0.0939 - val_accuracy: 0.9769 Epoch 10/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0289 - accuracy: 0.9913 - val_loss: 0.0830 - val_accuracy: 0.9800 Epoch 11/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0235 - accuracy: 0.9930 - val_loss: 0.0899 - val_accuracy: 0.9802 Epoch 12/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0249 - accuracy: 0.9930 - val_loss: 0.1014 - val_accuracy: 0.9782 Epoch 13/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0198 - accuracy: 0.9944 - val_loss: 0.0918 - val_accuracy: 0.9796 Epoch 14/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0182 - accuracy: 0.9949 - val_loss: 0.0872 - val_accuracy: 0.9813 Epoch 15/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0207 - accuracy: 0.9939 - val_loss: 0.0792 - val_accuracy: 0.9824 Epoch 16/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9961 - val_loss: 0.1060 - val_accuracy: 0.9814 Epoch 17/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0183 - accuracy: 0.9953 - val_loss: 0.1247 - val_accuracy: 0.9768 Epoch 18/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0147 - accuracy: 0.9959 - val_loss: 0.1050 - val_accuracy: 0.9797 Epoch 19/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0157 - accuracy: 0.9957 - val_loss: 0.1076 - val_accuracy: 0.9803 Epoch 20/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9966 - val_loss: 0.0978 - val_accuracy: 0.9826 Epoch 21/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0097 - accuracy: 0.9972 - val_loss: 0.1301 - val_accuracy: 0.9801 Epoch 22/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9969 - val_loss: 0.1251 - val_accuracy: 0.9797 Epoch 23/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0148 - accuracy: 0.9960 - val_loss: 0.1114 - val_accuracy: 0.9808 Epoch 24/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0093 - accuracy: 0.9974 - val_loss: 0.0999 - val_accuracy: 0.9823 Epoch 25/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9967 - val_loss: 0.1153 - val_accuracy: 0.9833 Epoch 26/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0115 - accuracy: 0.9970 - val_loss: 0.1006 - val_accuracy: 0.9829 Epoch 27/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0095 - accuracy: 0.9975 - val_loss: 0.1098 - val_accuracy: 0.9837 Epoch 28/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0103 - accuracy: 0.9973 - val_loss: 0.1225 - val_accuracy: 0.9818 Epoch 29/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0128 - accuracy: 0.9969 - val_loss: 0.1125 - val_accuracy: 0.9800 Epoch 30/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0081 - accuracy: 0.9977 - val_loss: 0.1246 - val_accuracy: 0.9827 Epoch 31/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0059 - accuracy: 0.9985 - val_loss: 0.1051 - val_accuracy: 0.9823 Epoch 32/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0079 - accuracy: 0.9981 - val_loss: 0.1333 - val_accuracy: 0.9805 Epoch 33/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9974 - val_loss: 0.1216 - val_accuracy: 0.9819 Epoch 34/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0079 - accuracy: 0.9981 - val_loss: 0.1232 - val_accuracy: 0.9811 Epoch 35/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0102 - accuracy: 0.9974 - val_loss: 0.1330 - val_accuracy: 0.9829 Epoch 36/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0087 - accuracy: 0.9980 - val_loss: 0.1052 - val_accuracy: 0.9824 Epoch 37/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0038 - accuracy: 0.9989 - val_loss: 0.1129 - val_accuracy: 0.9860 Epoch 38/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0018 - accuracy: 0.9996 - val_loss: 0.1258 - val_accuracy: 0.9838 Epoch 39/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0110 - accuracy: 0.9978 - val_loss: 0.1168 - val_accuracy: 0.9816 Epoch 40/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0106 - accuracy: 0.9973 - val_loss: 0.1050 - val_accuracy: 0.9865 Epoch 41/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0080 - accuracy: 0.9981 - val_loss: 0.1016 - val_accuracy: 0.9845 Epoch 42/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0076 - accuracy: 0.9981 - val_loss: 0.1251 - val_accuracy: 0.9806 Epoch 43/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0073 - accuracy: 0.9979 - val_loss: 0.1034 - val_accuracy: 0.9838 Epoch 44/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0064 - accuracy: 0.9981 - val_loss: 0.1102 - val_accuracy: 0.9836 Epoch 45/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0077 - accuracy: 0.9982 - val_loss: 0.1574 - val_accuracy: 0.9819 Epoch 46/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0084 - accuracy: 0.9983 - val_loss: 0.1013 - val_accuracy: 0.9835 Epoch 47/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0056 - accuracy: 0.9986 - val_loss: 0.1262 - val_accuracy: 0.9831 Epoch 48/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0028 - accuracy: 0.9991 - val_loss: 0.1280 - val_accuracy: 0.9858 Epoch 49/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0026 - accuracy: 0.9992 - val_loss: 0.1500 - val_accuracy: 0.9839 Epoch 50/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0081 - accuracy: 0.9983 - val_loss: 0.1113 - val_accuracy: 0.9843 Epoch 51/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0051 - accuracy: 0.9987 - val_loss: 0.1256 - val_accuracy: 0.9835 Epoch 52/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0062 - accuracy: 0.9988 - val_loss: 0.1755 - val_accuracy: 0.9819 Epoch 53/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9976 - val_loss: 0.0993 - val_accuracy: 0.9847 Epoch 54/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0020 - accuracy: 0.9995 - val_loss: 0.1082 - val_accuracy: 0.9849 Epoch 55/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0051 - accuracy: 0.9988 - val_loss: 0.1262 - val_accuracy: 0.9820 Epoch 56/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9987 - val_loss: 0.1302 - val_accuracy: 0.9846 Epoch 57/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0052 - accuracy: 0.9989 - val_loss: 0.2304 - val_accuracy: 0.9826 Epoch 58/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0082 - accuracy: 0.9982 - val_loss: 0.1203 - val_accuracy: 0.9845 Epoch 59/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0052 - accuracy: 0.9989 - val_loss: 0.1245 - val_accuracy: 0.9834 Epoch 60/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0045 - accuracy: 0.9992 - val_loss: 0.1285 - val_accuracy: 0.9855 Epoch 61/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0037 - accuracy: 0.9990 - val_loss: 0.1615 - val_accuracy: 0.9840 Epoch 62/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0076 - accuracy: 0.9981 - val_loss: 0.1259 - val_accuracy: 0.9809 Epoch 63/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0062 - accuracy: 0.9987 - val_loss: 0.1147 - val_accuracy: 0.9837 Epoch 64/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0021 - accuracy: 0.9996 - val_loss: 0.1436 - val_accuracy: 0.9837 Epoch 65/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0033 - accuracy: 0.9994 - val_loss: 0.1407 - val_accuracy: 0.9833 Epoch 66/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0057 - accuracy: 0.9987 - val_loss: 0.1416 - val_accuracy: 0.9840 Epoch 67/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0077 - accuracy: 0.9983 - val_loss: 0.1277 - val_accuracy: 0.9836 Epoch 68/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0042 - accuracy: 0.9992 - val_loss: 0.1467 - val_accuracy: 0.9824 Epoch 69/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0035 - accuracy: 0.9992 - val_loss: 0.1756 - val_accuracy: 0.9821 Epoch 70/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0063 - accuracy: 0.9984 - val_loss: 0.1143 - val_accuracy: 0.9849 Epoch 71/200 469/469 [==============================] - 2s 4ms/step - loss: 5.5770e-04 - accuracy: 0.9999 - val_loss: 0.1566 - val_accuracy: 0.9833 Epoch 72/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0031 - accuracy: 0.9995 - val_loss: 0.1597 - val_accuracy: 0.9826 Epoch 73/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0093 - accuracy: 0.9980 - val_loss: 0.1372 - val_accuracy: 0.9838 Epoch 74/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0070 - accuracy: 0.9987 - val_loss: 0.1707 - val_accuracy: 0.9810 Epoch 75/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0051 - accuracy: 0.9988 - val_loss: 0.1386 - val_accuracy: 0.9829 Epoch 76/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0042 - accuracy: 0.9993 - val_loss: 0.1187 - val_accuracy: 0.9834 Epoch 77/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0031 - accuracy: 0.9994 - val_loss: 0.1860 - val_accuracy: 0.9810 Epoch 78/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0046 - accuracy: 0.9991 - val_loss: 0.1719 - val_accuracy: 0.9838 Epoch 79/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0055 - accuracy: 0.9991 - val_loss: 0.1552 - val_accuracy: 0.9860 Epoch 80/200 469/469 [==============================] - 2s 4ms/step - loss: 4.1390e-04 - accuracy: 0.9999 - val_loss: 0.1969 - val_accuracy: 0.9855 Epoch 81/200 469/469 [==============================] - 2s 3ms/step - loss: 5.9587e-06 - accuracy: 1.0000 - val_loss: 0.2080 - val_accuracy: 0.9859 Epoch 82/200 469/469 [==============================] - 2s 4ms/step - loss: 1.2535e-06 - accuracy: 1.0000 - val_loss: 0.2221 - val_accuracy: 0.9859 Epoch 83/200 469/469 [==============================] - 2s 3ms/step - loss: 5.5130e-07 - accuracy: 1.0000 - val_loss: 0.2363 - val_accuracy: 0.9858 Epoch 84/200 469/469 [==============================] - 2s 3ms/step - loss: 2.7830e-07 - accuracy: 1.0000 - val_loss: 0.2480 - val_accuracy: 0.9858 Epoch 85/200 469/469 [==============================] - 2s 4ms/step - loss: 1.6186e-07 - accuracy: 1.0000 - val_loss: 0.2579 - val_accuracy: 0.9859 Epoch 86/200 469/469 [==============================] - 2s 3ms/step - loss: 1.0094e-07 - accuracy: 1.0000 - val_loss: 0.2670 - val_accuracy: 0.9860 Epoch 87/200 469/469 [==============================] - 2s 3ms/step - loss: 6.5931e-08 - accuracy: 1.0000 - val_loss: 0.2749 - val_accuracy: 0.9860 Epoch 88/200 469/469 [==============================] - 2s 4ms/step - loss: 4.4614e-08 - accuracy: 1.0000 - val_loss: 0.2820 - val_accuracy: 0.9860 Epoch 89/200 469/469 [==============================] - 2s 4ms/step - loss: 3.0950e-08 - accuracy: 1.0000 - val_loss: 0.2892 - val_accuracy: 0.9860 Epoch 90/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0931e-08 - accuracy: 1.0000 - val_loss: 0.2952 - val_accuracy: 0.9861 Epoch 91/200 469/469 [==============================] - 2s 4ms/step - loss: 1.5086e-08 - accuracy: 1.0000 - val_loss: 0.3005 - val_accuracy: 0.9861 Epoch 92/200 469/469 [==============================] - 2s 4ms/step - loss: 1.1146e-08 - accuracy: 1.0000 - val_loss: 0.3054 - val_accuracy: 0.9860 Epoch 93/200 469/469 [==============================] - 2s 4ms/step - loss: 8.4359e-09 - accuracy: 1.0000 - val_loss: 0.3099 - val_accuracy: 0.9860 Epoch 94/200 469/469 [==============================] - 2s 4ms/step - loss: 6.4889e-09 - accuracy: 1.0000 - val_loss: 0.3142 - val_accuracy: 0.9860 Epoch 95/200 469/469 [==============================] - 2s 3ms/step - loss: 5.0485e-09 - accuracy: 1.0000 - val_loss: 0.3183 - val_accuracy: 0.9859 Epoch 96/200 469/469 [==============================] - 2s 4ms/step - loss: 3.9518e-09 - accuracy: 1.0000 - val_loss: 0.3223 - val_accuracy: 0.9859 Epoch 97/200 469/469 [==============================] - 2s 3ms/step - loss: 3.0915e-09 - accuracy: 1.0000 - val_loss: 0.3259 - val_accuracy: 0.9859 Epoch 98/200 469/469 [==============================] - 2s 4ms/step - loss: 2.4160e-09 - accuracy: 1.0000 - val_loss: 0.3298 - val_accuracy: 0.9860 Epoch 99/200 469/469 [==============================] - 2s 3ms/step - loss: 1.9193e-09 - accuracy: 1.0000 - val_loss: 0.3331 - val_accuracy: 0.9860 Epoch 100/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5438e-09 - accuracy: 1.0000 - val_loss: 0.3362 - val_accuracy: 0.9859 Epoch 101/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2338e-09 - accuracy: 1.0000 - val_loss: 0.3391 - val_accuracy: 0.9859 Epoch 102/200 469/469 [==============================] - 2s 3ms/step - loss: 9.8546e-10 - accuracy: 1.0000 - val_loss: 0.3420 - val_accuracy: 0.9859 Epoch 103/200 469/469 [==============================] - 2s 3ms/step - loss: 8.2652e-10 - accuracy: 1.0000 - val_loss: 0.3446 - val_accuracy: 0.9859 Epoch 104/200 469/469 [==============================] - 2s 4ms/step - loss: 6.9539e-10 - accuracy: 1.0000 - val_loss: 0.3470 - val_accuracy: 0.9859 Epoch 105/200 469/469 [==============================] - 2s 3ms/step - loss: 5.8015e-10 - accuracy: 1.0000 - val_loss: 0.3493 - val_accuracy: 0.9859 Epoch 106/200 469/469 [==============================] - 2s 3ms/step - loss: 4.7684e-10 - accuracy: 1.0000 - val_loss: 0.3514 - val_accuracy: 0.9859 Epoch 107/200 469/469 [==============================] - 2s 3ms/step - loss: 4.1922e-10 - accuracy: 1.0000 - val_loss: 0.3534 - val_accuracy: 0.9859 Epoch 108/200 469/469 [==============================] - 2s 4ms/step - loss: 3.5365e-10 - accuracy: 1.0000 - val_loss: 0.3553 - val_accuracy: 0.9859 Epoch 109/200 469/469 [==============================] - 2s 4ms/step - loss: 3.0001e-10 - accuracy: 1.0000 - val_loss: 0.3571 - val_accuracy: 0.9859 Epoch 110/200 469/469 [==============================] - 2s 4ms/step - loss: 2.5233e-10 - accuracy: 1.0000 - val_loss: 0.3588 - val_accuracy: 0.9859 Epoch 111/200 469/469 [==============================] - 2s 3ms/step - loss: 2.1656e-10 - accuracy: 1.0000 - val_loss: 0.3605 - val_accuracy: 0.9859 Epoch 112/200 469/469 [==============================] - 2s 4ms/step - loss: 1.7285e-10 - accuracy: 1.0000 - val_loss: 0.3620 - val_accuracy: 0.9859 Epoch 113/200 469/469 [==============================] - 2s 4ms/step - loss: 1.5299e-10 - accuracy: 1.0000 - val_loss: 0.3635 - val_accuracy: 0.9859 Epoch 114/200 469/469 [==============================] - 2s 4ms/step - loss: 1.3312e-10 - accuracy: 1.0000 - val_loss: 0.3649 - val_accuracy: 0.9859 Epoch 115/200 469/469 [==============================] - 2s 4ms/step - loss: 1.0928e-10 - accuracy: 1.0000 - val_loss: 0.3663 - val_accuracy: 0.9859 Epoch 116/200 469/469 [==============================] - 2s 4ms/step - loss: 9.7354e-11 - accuracy: 1.0000 - val_loss: 0.3676 - val_accuracy: 0.9859 Epoch 117/200 469/469 [==============================] - 2s 4ms/step - loss: 8.3446e-11 - accuracy: 1.0000 - val_loss: 0.3689 - val_accuracy: 0.9859 Epoch 118/200 469/469 [==============================] - 2s 4ms/step - loss: 6.5565e-11 - accuracy: 1.0000 - val_loss: 0.3701 - val_accuracy: 0.9859 Epoch 119/200 469/469 [==============================] - 2s 3ms/step - loss: 4.7684e-11 - accuracy: 1.0000 - val_loss: 0.3713 - val_accuracy: 0.9859 Epoch 120/200 469/469 [==============================] - 2s 3ms/step - loss: 4.1723e-11 - accuracy: 1.0000 - val_loss: 0.3724 - val_accuracy: 0.9859 Epoch 121/200 469/469 [==============================] - 2s 3ms/step - loss: 3.1789e-11 - accuracy: 1.0000 - val_loss: 0.3735 - val_accuracy: 0.9859 Epoch 122/200 469/469 [==============================] - 2s 3ms/step - loss: 2.9802e-11 - accuracy: 1.0000 - val_loss: 0.3746 - val_accuracy: 0.9859 Epoch 123/200 469/469 [==============================] - 2s 4ms/step - loss: 2.1855e-11 - accuracy: 1.0000 - val_loss: 0.3756 - val_accuracy: 0.9859 Epoch 124/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5895e-11 - accuracy: 1.0000 - val_loss: 0.3765 - val_accuracy: 0.9859 Epoch 125/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1921e-11 - accuracy: 1.0000 - val_loss: 0.3775 - val_accuracy: 0.9858 Epoch 126/200 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-12 - accuracy: 1.0000 - val_loss: 0.3783 - val_accuracy: 0.9858 Epoch 127/200 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-12 - accuracy: 1.0000 - val_loss: 0.3792 - val_accuracy: 0.9859 Epoch 128/200 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-12 - accuracy: 1.0000 - val_loss: 0.3800 - val_accuracy: 0.9859 Epoch 129/200 469/469 [==============================] - 2s 4ms/step - loss: 7.9473e-12 - accuracy: 1.0000 - val_loss: 0.3807 - val_accuracy: 0.9859 Epoch 130/200 469/469 [==============================] - 2s 4ms/step - loss: 5.9605e-12 - accuracy: 1.0000 - val_loss: 0.3814 - val_accuracy: 0.9859 Epoch 131/200 469/469 [==============================] - 2s 4ms/step - loss: 5.9605e-12 - accuracy: 1.0000 - val_loss: 0.3821 - val_accuracy: 0.9859 Epoch 132/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3827 - val_accuracy: 0.9858 Epoch 133/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3834 - val_accuracy: 0.9858 Epoch 134/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3839 - val_accuracy: 0.9858 Epoch 135/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3845 - val_accuracy: 0.9857 Epoch 136/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3850 - val_accuracy: 0.9857 Epoch 137/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3855 - val_accuracy: 0.9857 Epoch 138/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3860 - val_accuracy: 0.9857 Epoch 139/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3864 - val_accuracy: 0.9857 Epoch 140/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3869 - val_accuracy: 0.9857 Epoch 141/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3873 - val_accuracy: 0.9857 Epoch 142/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3877 - val_accuracy: 0.9857 Epoch 143/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3881 - val_accuracy: 0.9857 Epoch 144/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3885 - val_accuracy: 0.9857 Epoch 145/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3889 - val_accuracy: 0.9857 Epoch 146/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3892 - val_accuracy: 0.9857 Epoch 147/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3896 - val_accuracy: 0.9857 Epoch 148/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3899 - val_accuracy: 0.9857 Epoch 149/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3902 - val_accuracy: 0.9857 Epoch 150/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3905 - val_accuracy: 0.9857 Epoch 151/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3908 - val_accuracy: 0.9857 Epoch 152/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3911 - val_accuracy: 0.9857 Epoch 153/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3914 - val_accuracy: 0.9857 Epoch 154/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3917 - val_accuracy: 0.9857 Epoch 155/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3920 - val_accuracy: 0.9857 Epoch 156/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3922 - val_accuracy: 0.9857 Epoch 157/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3925 - val_accuracy: 0.9857 Epoch 158/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3927 - val_accuracy: 0.9857 Epoch 159/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3930 - val_accuracy: 0.9857 Epoch 160/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3932 - val_accuracy: 0.9857 Epoch 161/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3935 - val_accuracy: 0.9857 Epoch 162/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3937 - val_accuracy: 0.9858 Epoch 163/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3939 - val_accuracy: 0.9858 Epoch 164/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3942 - val_accuracy: 0.9858 Epoch 165/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3944 - val_accuracy: 0.9858 Epoch 166/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3946 - val_accuracy: 0.9858 Epoch 167/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3948 - val_accuracy: 0.9858 Epoch 168/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3950 - val_accuracy: 0.9858 Epoch 169/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3952 - val_accuracy: 0.9858 Epoch 170/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3954 - val_accuracy: 0.9858 Epoch 171/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3956 - val_accuracy: 0.9858 Epoch 172/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3958 - val_accuracy: 0.9858 Epoch 173/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3960 - val_accuracy: 0.9858 Epoch 174/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3962 - val_accuracy: 0.9858 Epoch 175/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3963 - val_accuracy: 0.9858 Epoch 176/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3965 - val_accuracy: 0.9858 Epoch 177/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3967 - val_accuracy: 0.9858 Epoch 178/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3969 - val_accuracy: 0.9858 Epoch 179/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3970 - val_accuracy: 0.9858 Epoch 180/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3972 - val_accuracy: 0.9858 Epoch 181/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3974 - val_accuracy: 0.9858 Epoch 182/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3975 - val_accuracy: 0.9858 Epoch 183/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3977 - val_accuracy: 0.9858 Epoch 184/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3979 - val_accuracy: 0.9858 Epoch 185/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3980 - val_accuracy: 0.9858 Epoch 186/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3982 - val_accuracy: 0.9858 Epoch 187/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3983 - val_accuracy: 0.9858 Epoch 188/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3985 - val_accuracy: 0.9858 Epoch 189/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3986 - val_accuracy: 0.9858 Epoch 190/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3988 - val_accuracy: 0.9858 Epoch 191/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3989 - val_accuracy: 0.9858 Epoch 192/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3990 - val_accuracy: 0.9858 Epoch 193/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3992 - val_accuracy: 0.9858 Epoch 194/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3993 - val_accuracy: 0.9858 Epoch 195/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3995 - val_accuracy: 0.9858 Epoch 196/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3996 - val_accuracy: 0.9858 Epoch 197/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3997 - val_accuracy: 0.9858 Epoch 198/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3998 - val_accuracy: 0.9858 Epoch 199/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4000 - val_accuracy: 0.9858 Epoch 200/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4001 - val_accuracy: 0.9858
(d) Activation function: ReLU; initialization: Xavier initializer¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.GlorotNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
#tf.keras.layers.Dense(512,kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_6" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_6 (Flatten) (None, 784) 0 _________________________________________________________________ dense_36 (Dense) (None, 512) 401920 _________________________________________________________________ dense_37 (Dense) (None, 512) 262656 _________________________________________________________________ dense_38 (Dense) (None, 512) 262656 _________________________________________________________________ dense_39 (Dense) (None, 512) 262656 _________________________________________________________________ dense_40 (Dense) (None, 512) 262656 _________________________________________________________________ dense_41 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam() #learning_rate=1.0 for SGD
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
historya3 = model.fit(X_train, y_train, batch_size=128, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 469/469 [==============================] - 2s 4ms/step - loss: 0.2291 - accuracy: 0.9307 - val_loss: 0.1238 - val_accuracy: 0.9626 Epoch 2/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0967 - accuracy: 0.9708 - val_loss: 0.1124 - val_accuracy: 0.9680 Epoch 3/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0650 - accuracy: 0.9804 - val_loss: 0.0921 - val_accuracy: 0.9751 Epoch 4/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0542 - accuracy: 0.9841 - val_loss: 0.0896 - val_accuracy: 0.9764 Epoch 5/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0432 - accuracy: 0.9869 - val_loss: 0.0926 - val_accuracy: 0.9761 Epoch 6/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0367 - accuracy: 0.9890 - val_loss: 0.0720 - val_accuracy: 0.9811 Epoch 7/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0320 - accuracy: 0.9906 - val_loss: 0.0765 - val_accuracy: 0.9798 Epoch 8/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0304 - accuracy: 0.9911 - val_loss: 0.0933 - val_accuracy: 0.9769 Epoch 9/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0254 - accuracy: 0.9925 - val_loss: 0.0913 - val_accuracy: 0.9799 Epoch 10/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0246 - accuracy: 0.9929 - val_loss: 0.1032 - val_accuracy: 0.9768 Epoch 11/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0222 - accuracy: 0.9935 - val_loss: 0.1019 - val_accuracy: 0.9807 Epoch 12/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0200 - accuracy: 0.9945 - val_loss: 0.1137 - val_accuracy: 0.9773 Epoch 13/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0205 - accuracy: 0.9942 - val_loss: 0.0887 - val_accuracy: 0.9817 Epoch 14/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0162 - accuracy: 0.9952 - val_loss: 0.1121 - val_accuracy: 0.9807 Epoch 15/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0145 - accuracy: 0.9959 - val_loss: 0.0970 - val_accuracy: 0.9818 Epoch 16/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0162 - accuracy: 0.9960 - val_loss: 0.1034 - val_accuracy: 0.9803 Epoch 17/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0154 - accuracy: 0.9960 - val_loss: 0.0879 - val_accuracy: 0.9826 Epoch 18/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0168 - accuracy: 0.9953 - val_loss: 0.0903 - val_accuracy: 0.9841 Epoch 19/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0146 - accuracy: 0.9963 - val_loss: 0.0978 - val_accuracy: 0.9828 Epoch 20/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0140 - accuracy: 0.9961 - val_loss: 0.0921 - val_accuracy: 0.9826 Epoch 21/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9962 - val_loss: 0.1006 - val_accuracy: 0.9816 Epoch 22/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0099 - accuracy: 0.9974 - val_loss: 0.1134 - val_accuracy: 0.9813 Epoch 23/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0147 - accuracy: 0.9959 - val_loss: 0.0989 - val_accuracy: 0.9838 Epoch 24/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0125 - accuracy: 0.9971 - val_loss: 0.1140 - val_accuracy: 0.9834 Epoch 25/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0080 - accuracy: 0.9979 - val_loss: 0.1232 - val_accuracy: 0.9793 Epoch 26/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0102 - accuracy: 0.9975 - val_loss: 0.1224 - val_accuracy: 0.9801 Epoch 27/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0121 - accuracy: 0.9967 - val_loss: 0.1040 - val_accuracy: 0.9800 Epoch 28/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0121 - accuracy: 0.9967 - val_loss: 0.1114 - val_accuracy: 0.9818 Epoch 29/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0090 - accuracy: 0.9980 - val_loss: 0.0949 - val_accuracy: 0.9854 Epoch 30/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0104 - accuracy: 0.9975 - val_loss: 0.1287 - val_accuracy: 0.9817 Epoch 31/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0109 - accuracy: 0.9975 - val_loss: 0.1123 - val_accuracy: 0.9809 Epoch 32/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9972 - val_loss: 0.1041 - val_accuracy: 0.9835 Epoch 33/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0091 - accuracy: 0.9981 - val_loss: 0.1234 - val_accuracy: 0.9838 Epoch 34/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0066 - accuracy: 0.9983 - val_loss: 0.1298 - val_accuracy: 0.9836 Epoch 35/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0073 - accuracy: 0.9983 - val_loss: 0.1012 - val_accuracy: 0.9837 Epoch 36/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0066 - accuracy: 0.9984 - val_loss: 0.1138 - val_accuracy: 0.9824 Epoch 37/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0056 - accuracy: 0.9987 - val_loss: 0.1367 - val_accuracy: 0.9824 Epoch 38/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0130 - accuracy: 0.9970 - val_loss: 0.1444 - val_accuracy: 0.9815 Epoch 39/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0080 - accuracy: 0.9981 - val_loss: 0.1229 - val_accuracy: 0.9834 Epoch 40/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0091 - accuracy: 0.9981 - val_loss: 0.1500 - val_accuracy: 0.9838 Epoch 41/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0079 - accuracy: 0.9984 - val_loss: 0.1131 - val_accuracy: 0.9824 Epoch 42/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0111 - accuracy: 0.9977 - val_loss: 0.1241 - val_accuracy: 0.9814 Epoch 43/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0048 - accuracy: 0.9988 - val_loss: 0.1374 - val_accuracy: 0.9824 Epoch 44/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0074 - accuracy: 0.9981 - val_loss: 0.1338 - val_accuracy: 0.9828 Epoch 45/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0117 - accuracy: 0.9976 - val_loss: 0.1270 - val_accuracy: 0.9829 Epoch 46/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0073 - accuracy: 0.9984 - val_loss: 0.1266 - val_accuracy: 0.9812 Epoch 47/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0048 - accuracy: 0.9990 - val_loss: 0.1244 - val_accuracy: 0.9819 Epoch 48/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0071 - accuracy: 0.9984 - val_loss: 0.1085 - val_accuracy: 0.9834 Epoch 49/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0076 - accuracy: 0.9983 - val_loss: 0.1432 - val_accuracy: 0.9831 Epoch 50/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0056 - accuracy: 0.9988 - val_loss: 0.1296 - val_accuracy: 0.9840 Epoch 51/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0078 - accuracy: 0.9982 - val_loss: 0.1483 - val_accuracy: 0.9838 Epoch 52/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0061 - accuracy: 0.9986 - val_loss: 0.1252 - val_accuracy: 0.9840 Epoch 53/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0085 - accuracy: 0.9984 - val_loss: 0.1434 - val_accuracy: 0.9849 Epoch 54/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9988 - val_loss: 0.1637 - val_accuracy: 0.9851 Epoch 55/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0060 - accuracy: 0.9988 - val_loss: 0.1263 - val_accuracy: 0.9838 Epoch 56/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0025 - accuracy: 0.9995 - val_loss: 0.1191 - val_accuracy: 0.9855 Epoch 57/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0060 - accuracy: 0.9988 - val_loss: 0.1647 - val_accuracy: 0.9816 Epoch 58/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0089 - accuracy: 0.9986 - val_loss: 0.1162 - val_accuracy: 0.9838 Epoch 59/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0062 - accuracy: 0.9986 - val_loss: 0.1210 - val_accuracy: 0.9825 Epoch 60/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0032 - accuracy: 0.9993 - val_loss: 0.1374 - val_accuracy: 0.9834 Epoch 61/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0047 - accuracy: 0.9991 - val_loss: 0.1455 - val_accuracy: 0.9828 Epoch 62/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0103 - accuracy: 0.9984 - val_loss: 0.1475 - val_accuracy: 0.9813 Epoch 63/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0074 - accuracy: 0.9986 - val_loss: 0.1164 - val_accuracy: 0.9836 Epoch 64/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0037 - accuracy: 0.9992 - val_loss: 0.1318 - val_accuracy: 0.9830 Epoch 65/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0107 - accuracy: 0.9980 - val_loss: 0.1453 - val_accuracy: 0.9823 Epoch 66/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0049 - accuracy: 0.9990 - val_loss: 0.1541 - val_accuracy: 0.9838 Epoch 67/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0033 - accuracy: 0.9992 - val_loss: 0.1382 - val_accuracy: 0.9832 Epoch 68/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0050 - accuracy: 0.9991 - val_loss: 0.1491 - val_accuracy: 0.9837 Epoch 69/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0017 - accuracy: 0.9996 - val_loss: 0.1863 - val_accuracy: 0.9836 Epoch 70/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0061 - accuracy: 0.9985 - val_loss: 0.2057 - val_accuracy: 0.9807 Epoch 71/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0011 - accuracy: 0.9997 - val_loss: 0.1716 - val_accuracy: 0.9838 Epoch 72/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0079 - accuracy: 0.9987 - val_loss: 0.1275 - val_accuracy: 0.9843 Epoch 73/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9983 - val_loss: 0.1441 - val_accuracy: 0.9830 Epoch 74/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0025 - accuracy: 0.9995 - val_loss: 0.1443 - val_accuracy: 0.9842 Epoch 75/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0065 - accuracy: 0.9988 - val_loss: 0.1519 - val_accuracy: 0.9840 Epoch 76/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0053 - accuracy: 0.9988 - val_loss: 0.1472 - val_accuracy: 0.9831 Epoch 77/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0072 - accuracy: 0.9987 - val_loss: 0.1461 - val_accuracy: 0.9837 Epoch 78/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0034 - accuracy: 0.9991 - val_loss: 0.1609 - val_accuracy: 0.9829 Epoch 79/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0019 - accuracy: 0.9995 - val_loss: 0.1836 - val_accuracy: 0.9846 Epoch 80/200 469/469 [==============================] - 2s 4ms/step - loss: 3.0217e-05 - accuracy: 1.0000 - val_loss: 0.1969 - val_accuracy: 0.9852 Epoch 81/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9975 - val_loss: 0.1494 - val_accuracy: 0.9829 Epoch 82/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0053 - accuracy: 0.9989 - val_loss: 0.1611 - val_accuracy: 0.9834 Epoch 83/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0023 - accuracy: 0.9994 - val_loss: 0.1456 - val_accuracy: 0.9852 Epoch 84/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0029 - accuracy: 0.9995 - val_loss: 0.1556 - val_accuracy: 0.9848 Epoch 85/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0110 - accuracy: 0.9978 - val_loss: 0.1469 - val_accuracy: 0.9822 Epoch 86/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0028 - accuracy: 0.9993 - val_loss: 0.1325 - val_accuracy: 0.9852 Epoch 87/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0032 - accuracy: 0.9995 - val_loss: 0.1256 - val_accuracy: 0.9843 Epoch 88/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0025 - accuracy: 0.9995 - val_loss: 0.1624 - val_accuracy: 0.9850 Epoch 89/200 469/469 [==============================] - 2s 4ms/step - loss: 6.8777e-04 - accuracy: 0.9999 - val_loss: 0.1632 - val_accuracy: 0.9852 Epoch 90/200 469/469 [==============================] - 2s 3ms/step - loss: 6.1643e-04 - accuracy: 0.9999 - val_loss: 0.1674 - val_accuracy: 0.9860 Epoch 91/200 469/469 [==============================] - 2s 4ms/step - loss: 1.7304e-05 - accuracy: 1.0000 - val_loss: 0.1812 - val_accuracy: 0.9859 Epoch 92/200 469/469 [==============================] - 2s 4ms/step - loss: 2.7191e-06 - accuracy: 1.0000 - val_loss: 0.2012 - val_accuracy: 0.9863 Epoch 93/200 469/469 [==============================] - 2s 3ms/step - loss: 3.4179e-07 - accuracy: 1.0000 - val_loss: 0.2250 - val_accuracy: 0.9861 Epoch 94/200 469/469 [==============================] - 2s 4ms/step - loss: 1.2425e-07 - accuracy: 1.0000 - val_loss: 0.2440 - val_accuracy: 0.9860 Epoch 95/200 469/469 [==============================] - 2s 4ms/step - loss: 5.8452e-08 - accuracy: 1.0000 - val_loss: 0.2571 - val_accuracy: 0.9861 Epoch 96/200 469/469 [==============================] - 2s 3ms/step - loss: 3.2900e-08 - accuracy: 1.0000 - val_loss: 0.2673 - val_accuracy: 0.9861 Epoch 97/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0553e-08 - accuracy: 1.0000 - val_loss: 0.2759 - val_accuracy: 0.9861 Epoch 98/200 469/469 [==============================] - 2s 3ms/step - loss: 1.3876e-08 - accuracy: 1.0000 - val_loss: 0.2835 - val_accuracy: 0.9860 Epoch 99/200 469/469 [==============================] - 2s 3ms/step - loss: 9.6817e-09 - accuracy: 1.0000 - val_loss: 0.2898 - val_accuracy: 0.9861 Epoch 100/200 469/469 [==============================] - 2s 3ms/step - loss: 7.1068e-09 - accuracy: 1.0000 - val_loss: 0.2958 - val_accuracy: 0.9862 Epoch 101/200 469/469 [==============================] - 2s 4ms/step - loss: 5.3207e-09 - accuracy: 1.0000 - val_loss: 0.3010 - val_accuracy: 0.9861 Epoch 102/200 469/469 [==============================] - 2s 3ms/step - loss: 4.0849e-09 - accuracy: 1.0000 - val_loss: 0.3060 - val_accuracy: 0.9861 Epoch 103/200 469/469 [==============================] - 2s 4ms/step - loss: 3.1928e-09 - accuracy: 1.0000 - val_loss: 0.3106 - val_accuracy: 0.9860 Epoch 104/200 469/469 [==============================] - 2s 4ms/step - loss: 2.4815e-09 - accuracy: 1.0000 - val_loss: 0.3150 - val_accuracy: 0.9861 Epoch 105/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9590e-09 - accuracy: 1.0000 - val_loss: 0.3192 - val_accuracy: 0.9862 Epoch 106/200 469/469 [==============================] - 2s 4ms/step - loss: 1.5597e-09 - accuracy: 1.0000 - val_loss: 0.3228 - val_accuracy: 0.9862 Epoch 107/200 469/469 [==============================] - 2s 4ms/step - loss: 1.2577e-09 - accuracy: 1.0000 - val_loss: 0.3267 - val_accuracy: 0.9862 Epoch 108/200 469/469 [==============================] - 2s 4ms/step - loss: 1.0093e-09 - accuracy: 1.0000 - val_loss: 0.3301 - val_accuracy: 0.9862 Epoch 109/200 469/469 [==============================] - 2s 3ms/step - loss: 8.3248e-10 - accuracy: 1.0000 - val_loss: 0.3334 - val_accuracy: 0.9862 Epoch 110/200 469/469 [==============================] - 2s 4ms/step - loss: 6.9340e-10 - accuracy: 1.0000 - val_loss: 0.3365 - val_accuracy: 0.9861 Epoch 111/200 469/469 [==============================] - 2s 4ms/step - loss: 5.8015e-10 - accuracy: 1.0000 - val_loss: 0.3393 - val_accuracy: 0.9861 Epoch 112/200 469/469 [==============================] - 2s 4ms/step - loss: 4.7882e-10 - accuracy: 1.0000 - val_loss: 0.3423 - val_accuracy: 0.9861 Epoch 113/200 469/469 [==============================] - 2s 3ms/step - loss: 3.9736e-10 - accuracy: 1.0000 - val_loss: 0.3450 - val_accuracy: 0.9861 Epoch 114/200 469/469 [==============================] - 2s 4ms/step - loss: 3.3776e-10 - accuracy: 1.0000 - val_loss: 0.3476 - val_accuracy: 0.9861 Epoch 115/200 469/469 [==============================] - 2s 4ms/step - loss: 2.7418e-10 - accuracy: 1.0000 - val_loss: 0.3502 - val_accuracy: 0.9860 Epoch 116/200 469/469 [==============================] - 2s 3ms/step - loss: 2.4041e-10 - accuracy: 1.0000 - val_loss: 0.3528 - val_accuracy: 0.9859 Epoch 117/200 469/469 [==============================] - 2s 3ms/step - loss: 2.0266e-10 - accuracy: 1.0000 - val_loss: 0.3552 - val_accuracy: 0.9860 Epoch 118/200 469/469 [==============================] - 2s 4ms/step - loss: 1.7087e-10 - accuracy: 1.0000 - val_loss: 0.3574 - val_accuracy: 0.9860 Epoch 119/200 469/469 [==============================] - 2s 4ms/step - loss: 1.4901e-10 - accuracy: 1.0000 - val_loss: 0.3596 - val_accuracy: 0.9860 Epoch 120/200 469/469 [==============================] - 2s 4ms/step - loss: 1.2318e-10 - accuracy: 1.0000 - val_loss: 0.3619 - val_accuracy: 0.9860 Epoch 121/200 469/469 [==============================] - 2s 3ms/step - loss: 9.7354e-11 - accuracy: 1.0000 - val_loss: 0.3642 - val_accuracy: 0.9860 Epoch 122/200 469/469 [==============================] - 2s 4ms/step - loss: 7.9473e-11 - accuracy: 1.0000 - val_loss: 0.3665 - val_accuracy: 0.9860 Epoch 123/200 469/469 [==============================] - 2s 3ms/step - loss: 6.5565e-11 - accuracy: 1.0000 - val_loss: 0.3687 - val_accuracy: 0.9860 Epoch 124/200 469/469 [==============================] - 2s 3ms/step - loss: 5.5631e-11 - accuracy: 1.0000 - val_loss: 0.3706 - val_accuracy: 0.9860 Epoch 125/200 469/469 [==============================] - 2s 4ms/step - loss: 4.7684e-11 - accuracy: 1.0000 - val_loss: 0.3725 - val_accuracy: 0.9860 Epoch 126/200 469/469 [==============================] - 2s 3ms/step - loss: 3.3776e-11 - accuracy: 1.0000 - val_loss: 0.3742 - val_accuracy: 0.9860 Epoch 127/200 469/469 [==============================] - 2s 4ms/step - loss: 2.5829e-11 - accuracy: 1.0000 - val_loss: 0.3761 - val_accuracy: 0.9859 Epoch 128/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-11 - accuracy: 1.0000 - val_loss: 0.3779 - val_accuracy: 0.9859 Epoch 129/200 469/469 [==============================] - 2s 4ms/step - loss: 1.3908e-11 - accuracy: 1.0000 - val_loss: 0.3796 - val_accuracy: 0.9859 Epoch 130/200 469/469 [==============================] - 2s 4ms/step - loss: 1.3908e-11 - accuracy: 1.0000 - val_loss: 0.3811 - val_accuracy: 0.9859 Epoch 131/200 469/469 [==============================] - 2s 4ms/step - loss: 1.1921e-11 - accuracy: 1.0000 - val_loss: 0.3826 - val_accuracy: 0.9859 Epoch 132/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1921e-11 - accuracy: 1.0000 - val_loss: 0.3839 - val_accuracy: 0.9859 Epoch 133/200 469/469 [==============================] - 2s 3ms/step - loss: 9.9341e-12 - accuracy: 1.0000 - val_loss: 0.3853 - val_accuracy: 0.9859 Epoch 134/200 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-12 - accuracy: 1.0000 - val_loss: 0.3865 - val_accuracy: 0.9858 Epoch 135/200 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-12 - accuracy: 1.0000 - val_loss: 0.3876 - val_accuracy: 0.9858 Epoch 136/200 469/469 [==============================] - 2s 3ms/step - loss: 9.9341e-12 - accuracy: 1.0000 - val_loss: 0.3887 - val_accuracy: 0.9858 Epoch 137/200 469/469 [==============================] - 2s 3ms/step - loss: 7.9473e-12 - accuracy: 1.0000 - val_loss: 0.3896 - val_accuracy: 0.9857 Epoch 138/200 469/469 [==============================] - 2s 3ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.3908 - val_accuracy: 0.9857 Epoch 139/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3918 - val_accuracy: 0.9857 Epoch 140/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3928 - val_accuracy: 0.9857 Epoch 141/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3938 - val_accuracy: 0.9857 Epoch 142/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3946 - val_accuracy: 0.9857 Epoch 143/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3955 - val_accuracy: 0.9857 Epoch 144/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3963 - val_accuracy: 0.9857 Epoch 145/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3970 - val_accuracy: 0.9857 Epoch 146/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3978 - val_accuracy: 0.9857 Epoch 147/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3985 - val_accuracy: 0.9857 Epoch 148/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3992 - val_accuracy: 0.9857 Epoch 149/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3998 - val_accuracy: 0.9857 Epoch 150/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4005 - val_accuracy: 0.9857 Epoch 151/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4011 - val_accuracy: 0.9857 Epoch 152/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4016 - val_accuracy: 0.9857 Epoch 153/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4022 - val_accuracy: 0.9857 Epoch 154/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4028 - val_accuracy: 0.9857 Epoch 155/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4033 - val_accuracy: 0.9857 Epoch 156/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4038 - val_accuracy: 0.9857 Epoch 157/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4043 - val_accuracy: 0.9857 Epoch 158/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4047 - val_accuracy: 0.9857 Epoch 159/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4052 - val_accuracy: 0.9857 Epoch 160/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4057 - val_accuracy: 0.9857 Epoch 161/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4061 - val_accuracy: 0.9857 Epoch 162/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4065 - val_accuracy: 0.9856 Epoch 163/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4069 - val_accuracy: 0.9856 Epoch 164/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4073 - val_accuracy: 0.9856 Epoch 165/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4077 - val_accuracy: 0.9856 Epoch 166/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4081 - val_accuracy: 0.9856 Epoch 167/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4085 - val_accuracy: 0.9856 Epoch 168/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4088 - val_accuracy: 0.9856 Epoch 169/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4092 - val_accuracy: 0.9856 Epoch 170/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4096 - val_accuracy: 0.9856 Epoch 171/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4099 - val_accuracy: 0.9856 Epoch 172/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4102 - val_accuracy: 0.9856 Epoch 173/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4106 - val_accuracy: 0.9856 Epoch 174/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4109 - val_accuracy: 0.9856 Epoch 175/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4112 - val_accuracy: 0.9856 Epoch 176/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4115 - val_accuracy: 0.9856 Epoch 177/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4118 - val_accuracy: 0.9856 Epoch 178/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4121 - val_accuracy: 0.9856 Epoch 179/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4124 - val_accuracy: 0.9856 Epoch 180/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4127 - val_accuracy: 0.9856 Epoch 181/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4130 - val_accuracy: 0.9856 Epoch 182/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4133 - val_accuracy: 0.9856 Epoch 183/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4136 - val_accuracy: 0.9856 Epoch 184/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4138 - val_accuracy: 0.9856 Epoch 185/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4141 - val_accuracy: 0.9856 Epoch 186/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4144 - val_accuracy: 0.9856 Epoch 187/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4146 - val_accuracy: 0.9856 Epoch 188/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4149 - val_accuracy: 0.9856 Epoch 189/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4151 - val_accuracy: 0.9856 Epoch 190/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4154 - val_accuracy: 0.9856 Epoch 191/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4156 - val_accuracy: 0.9856 Epoch 192/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4159 - val_accuracy: 0.9856 Epoch 193/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4161 - val_accuracy: 0.9856 Epoch 194/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4163 - val_accuracy: 0.9856 Epoch 195/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4166 - val_accuracy: 0.9856 Epoch 196/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4168 - val_accuracy: 0.9856 Epoch 197/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4170 - val_accuracy: 0.9856 Epoch 198/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4172 - val_accuracy: 0.9856 Epoch 199/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4174 - val_accuracy: 0.9856 Epoch 200/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.4176 - val_accuracy: 0.9856
(e) Activation function: ReLU; initialization: Kaiming He’s initializer¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.HeNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
#tf.keras.layers.Dense(512,kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(512,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_7" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_7 (Flatten) (None, 784) 0 _________________________________________________________________ dense_42 (Dense) (None, 512) 401920 _________________________________________________________________ dense_43 (Dense) (None, 512) 262656 _________________________________________________________________ dense_44 (Dense) (None, 512) 262656 _________________________________________________________________ dense_45 (Dense) (None, 512) 262656 _________________________________________________________________ dense_46 (Dense) (None, 512) 262656 _________________________________________________________________ dense_47 (Dense) (None, 10) 5130 ================================================================= Total params: 1,457,674 Trainable params: 1,457,674 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam() #learning_rate=1.0 for SGD
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"]) #categorical_crossentropy
historya4 = model.fit(X_train, y_train, batch_size=128, epochs=200, validation_data=(X_test, y_test))
Epoch 1/200 469/469 [==============================] - 2s 4ms/step - loss: 0.2093 - accuracy: 0.9358 - val_loss: 0.1174 - val_accuracy: 0.9653 Epoch 2/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0905 - accuracy: 0.9730 - val_loss: 0.0985 - val_accuracy: 0.9698 Epoch 3/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0646 - accuracy: 0.9798 - val_loss: 0.1142 - val_accuracy: 0.9695 Epoch 4/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0512 - accuracy: 0.9847 - val_loss: 0.0903 - val_accuracy: 0.9774 Epoch 5/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0438 - accuracy: 0.9865 - val_loss: 0.0935 - val_accuracy: 0.9761 Epoch 6/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0321 - accuracy: 0.9900 - val_loss: 0.1073 - val_accuracy: 0.9723 Epoch 7/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0302 - accuracy: 0.9909 - val_loss: 0.0922 - val_accuracy: 0.9771 Epoch 8/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0301 - accuracy: 0.9909 - val_loss: 0.0836 - val_accuracy: 0.9775 Epoch 9/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0254 - accuracy: 0.9925 - val_loss: 0.0914 - val_accuracy: 0.9759 Epoch 10/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0227 - accuracy: 0.9933 - val_loss: 0.0866 - val_accuracy: 0.9782 Epoch 11/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0246 - accuracy: 0.9928 - val_loss: 0.0697 - val_accuracy: 0.9821 Epoch 12/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0202 - accuracy: 0.9941 - val_loss: 0.0953 - val_accuracy: 0.9787 Epoch 13/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0194 - accuracy: 0.9944 - val_loss: 0.1082 - val_accuracy: 0.9764 Epoch 14/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0187 - accuracy: 0.9947 - val_loss: 0.1104 - val_accuracy: 0.9800 Epoch 15/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0155 - accuracy: 0.9955 - val_loss: 0.0843 - val_accuracy: 0.9824 Epoch 16/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0166 - accuracy: 0.9951 - val_loss: 0.0967 - val_accuracy: 0.9818 Epoch 17/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9965 - val_loss: 0.1141 - val_accuracy: 0.9781 Epoch 18/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9959 - val_loss: 0.1209 - val_accuracy: 0.9769 Epoch 19/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0177 - accuracy: 0.9954 - val_loss: 0.1097 - val_accuracy: 0.9798 Epoch 20/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0117 - accuracy: 0.9969 - val_loss: 0.1034 - val_accuracy: 0.9805 Epoch 21/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0116 - accuracy: 0.9969 - val_loss: 0.0970 - val_accuracy: 0.9815 Epoch 22/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0103 - accuracy: 0.9973 - val_loss: 0.0958 - val_accuracy: 0.9809 Epoch 23/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0148 - accuracy: 0.9958 - val_loss: 0.0971 - val_accuracy: 0.9800 Epoch 24/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0124 - accuracy: 0.9969 - val_loss: 0.1026 - val_accuracy: 0.9823 Epoch 25/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9968 - val_loss: 0.1028 - val_accuracy: 0.9840 Epoch 26/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0135 - accuracy: 0.9969 - val_loss: 0.1196 - val_accuracy: 0.9822 Epoch 27/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0113 - accuracy: 0.9971 - val_loss: 0.1103 - val_accuracy: 0.9827 Epoch 28/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0099 - accuracy: 0.9975 - val_loss: 0.1227 - val_accuracy: 0.9787 Epoch 29/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0107 - accuracy: 0.9974 - val_loss: 0.1318 - val_accuracy: 0.9819 Epoch 30/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0154 - accuracy: 0.9966 - val_loss: 0.1321 - val_accuracy: 0.9794 Epoch 31/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0110 - accuracy: 0.9975 - val_loss: 0.1303 - val_accuracy: 0.9803 Epoch 32/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0077 - accuracy: 0.9980 - val_loss: 0.1664 - val_accuracy: 0.9808 Epoch 33/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0104 - accuracy: 0.9977 - val_loss: 0.1329 - val_accuracy: 0.9827 Epoch 34/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9987 - val_loss: 0.1405 - val_accuracy: 0.9810 Epoch 35/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9972 - val_loss: 0.1255 - val_accuracy: 0.9849 Epoch 36/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0116 - accuracy: 0.9970 - val_loss: 0.1166 - val_accuracy: 0.9793 Epoch 37/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0106 - accuracy: 0.9972 - val_loss: 0.1248 - val_accuracy: 0.9836 Epoch 38/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0078 - accuracy: 0.9980 - val_loss: 0.1358 - val_accuracy: 0.9820 Epoch 39/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0098 - accuracy: 0.9979 - val_loss: 0.1356 - val_accuracy: 0.9815 Epoch 40/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0073 - accuracy: 0.9986 - val_loss: 0.1402 - val_accuracy: 0.9822 Epoch 41/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0095 - accuracy: 0.9979 - val_loss: 0.1297 - val_accuracy: 0.9843 Epoch 42/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0054 - accuracy: 0.9990 - val_loss: 0.1479 - val_accuracy: 0.9816 Epoch 43/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0121 - accuracy: 0.9972 - val_loss: 0.1274 - val_accuracy: 0.9841 Epoch 44/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0066 - accuracy: 0.9984 - val_loss: 0.1373 - val_accuracy: 0.9840 Epoch 45/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0061 - accuracy: 0.9985 - val_loss: 0.1484 - val_accuracy: 0.9842 Epoch 46/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0083 - accuracy: 0.9983 - val_loss: 0.1566 - val_accuracy: 0.9821 Epoch 47/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0089 - accuracy: 0.9981 - val_loss: 0.1253 - val_accuracy: 0.9832 Epoch 48/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0091 - accuracy: 0.9982 - val_loss: 0.1427 - val_accuracy: 0.9831 Epoch 49/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0067 - accuracy: 0.9986 - val_loss: 0.1455 - val_accuracy: 0.9839 Epoch 50/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0057 - accuracy: 0.9989 - val_loss: 0.1449 - val_accuracy: 0.9827 Epoch 51/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0084 - accuracy: 0.9982 - val_loss: 0.1672 - val_accuracy: 0.9833 Epoch 52/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0091 - accuracy: 0.9983 - val_loss: 0.1455 - val_accuracy: 0.9818 Epoch 53/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9975 - val_loss: 0.1809 - val_accuracy: 0.9785 Epoch 54/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0101 - accuracy: 0.9979 - val_loss: 0.1631 - val_accuracy: 0.9819 Epoch 55/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0076 - accuracy: 0.9982 - val_loss: 0.1181 - val_accuracy: 0.9818 Epoch 56/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0022 - accuracy: 0.9995 - val_loss: 0.1516 - val_accuracy: 0.9829 Epoch 57/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0056 - accuracy: 0.9988 - val_loss: 0.1584 - val_accuracy: 0.9814 Epoch 58/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0040 - accuracy: 0.9990 - val_loss: 0.1644 - val_accuracy: 0.9821 Epoch 59/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0040 - accuracy: 0.9993 - val_loss: 0.1794 - val_accuracy: 0.9799 Epoch 60/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9969 - val_loss: 0.1640 - val_accuracy: 0.9840 Epoch 61/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0049 - accuracy: 0.9988 - val_loss: 0.1561 - val_accuracy: 0.9844 Epoch 62/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0032 - accuracy: 0.9995 - val_loss: 0.2156 - val_accuracy: 0.9805 Epoch 63/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0073 - accuracy: 0.9984 - val_loss: 0.1427 - val_accuracy: 0.9842 Epoch 64/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0027 - accuracy: 0.9994 - val_loss: 0.1883 - val_accuracy: 0.9779 Epoch 65/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0048 - accuracy: 0.9991 - val_loss: 0.1504 - val_accuracy: 0.9825 Epoch 66/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0077 - accuracy: 0.9986 - val_loss: 0.1935 - val_accuracy: 0.9828 Epoch 67/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0068 - accuracy: 0.9985 - val_loss: 0.2054 - val_accuracy: 0.9800 Epoch 68/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9981 - val_loss: 0.2093 - val_accuracy: 0.9804 Epoch 69/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0072 - accuracy: 0.9987 - val_loss: 0.1978 - val_accuracy: 0.9821 Epoch 70/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0092 - accuracy: 0.9983 - val_loss: 0.1696 - val_accuracy: 0.9820 Epoch 71/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0024 - accuracy: 0.9994 - val_loss: 0.1917 - val_accuracy: 0.9807 Epoch 72/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0048 - accuracy: 0.9992 - val_loss: 0.1642 - val_accuracy: 0.9851 Epoch 73/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0053 - accuracy: 0.9991 - val_loss: 0.1578 - val_accuracy: 0.9843 Epoch 74/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0084 - accuracy: 0.9982 - val_loss: 0.1628 - val_accuracy: 0.9836 Epoch 75/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0073 - accuracy: 0.9985 - val_loss: 0.1574 - val_accuracy: 0.9813 Epoch 76/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0047 - accuracy: 0.9990 - val_loss: 0.2003 - val_accuracy: 0.9794 Epoch 77/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0067 - accuracy: 0.9989 - val_loss: 0.1624 - val_accuracy: 0.9841 Epoch 78/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0068 - accuracy: 0.9990 - val_loss: 0.1470 - val_accuracy: 0.9828 Epoch 79/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0057 - accuracy: 0.9985 - val_loss: 0.1542 - val_accuracy: 0.9840 Epoch 80/200 469/469 [==============================] - 2s 3ms/step - loss: 7.3968e-04 - accuracy: 0.9998 - val_loss: 0.1885 - val_accuracy: 0.9841 Epoch 81/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0050 - accuracy: 0.9991 - val_loss: 0.1721 - val_accuracy: 0.9810 Epoch 82/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0159 - accuracy: 0.9977 - val_loss: 0.1768 - val_accuracy: 0.9804 Epoch 83/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0071 - accuracy: 0.9987 - val_loss: 0.1396 - val_accuracy: 0.9848 Epoch 84/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0043 - accuracy: 0.9991 - val_loss: 0.1574 - val_accuracy: 0.9824 Epoch 85/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0028 - accuracy: 0.9994 - val_loss: 0.2658 - val_accuracy: 0.9803 Epoch 86/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0045 - accuracy: 0.9993 - val_loss: 0.1905 - val_accuracy: 0.9825 Epoch 87/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0066 - accuracy: 0.9994 - val_loss: 0.1600 - val_accuracy: 0.9832 Epoch 88/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0037 - accuracy: 0.9995 - val_loss: 0.2075 - val_accuracy: 0.9810 Epoch 89/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9988 - val_loss: 0.1558 - val_accuracy: 0.9822 Epoch 90/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0022 - accuracy: 0.9995 - val_loss: 0.1730 - val_accuracy: 0.9850 Epoch 91/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0101 - accuracy: 0.9985 - val_loss: 0.1607 - val_accuracy: 0.9847 Epoch 92/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0067 - accuracy: 0.9988 - val_loss: 0.2095 - val_accuracy: 0.9828 Epoch 93/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0030 - accuracy: 0.9995 - val_loss: 0.2073 - val_accuracy: 0.9841 Epoch 94/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0014 - accuracy: 0.9996 - val_loss: 0.2250 - val_accuracy: 0.9836 Epoch 95/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0065 - accuracy: 0.9990 - val_loss: 0.1527 - val_accuracy: 0.9822 Epoch 96/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0087 - accuracy: 0.9987 - val_loss: 0.1973 - val_accuracy: 0.9811 Epoch 97/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0055 - accuracy: 0.9988 - val_loss: 0.2609 - val_accuracy: 0.9826 Epoch 98/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9989 - val_loss: 0.2434 - val_accuracy: 0.9844 Epoch 99/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0056 - accuracy: 0.9990 - val_loss: 0.1665 - val_accuracy: 0.9847 Epoch 100/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0027 - accuracy: 0.9994 - val_loss: 0.1821 - val_accuracy: 0.9832 Epoch 101/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0012 - accuracy: 0.9996 - val_loss: 0.2074 - val_accuracy: 0.9841 Epoch 102/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9984 - val_loss: 0.2045 - val_accuracy: 0.9823 Epoch 103/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0083 - accuracy: 0.9988 - val_loss: 0.1885 - val_accuracy: 0.9842 Epoch 104/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0017 - accuracy: 0.9997 - val_loss: 0.1869 - val_accuracy: 0.9845 Epoch 105/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0017 - accuracy: 0.9996 - val_loss: 0.2240 - val_accuracy: 0.9832 Epoch 106/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0024 - accuracy: 0.9995 - val_loss: 0.2144 - val_accuracy: 0.9832 Epoch 107/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9994 - val_loss: 0.1877 - val_accuracy: 0.9824 Epoch 108/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0130 - accuracy: 0.9978 - val_loss: 0.2717 - val_accuracy: 0.9799 Epoch 109/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0038 - accuracy: 0.9992 - val_loss: 0.2312 - val_accuracy: 0.9828 Epoch 110/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0028 - accuracy: 0.9995 - val_loss: 0.2224 - val_accuracy: 0.9848 Epoch 111/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0021 - accuracy: 0.9996 - val_loss: 0.3081 - val_accuracy: 0.9829 Epoch 112/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0034 - accuracy: 0.9996 - val_loss: 0.2012 - val_accuracy: 0.9800 Epoch 113/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0045 - accuracy: 0.9990 - val_loss: 0.2208 - val_accuracy: 0.9840 Epoch 114/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0059 - accuracy: 0.9991 - val_loss: 0.2775 - val_accuracy: 0.9825 Epoch 115/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0038 - accuracy: 0.9991 - val_loss: 0.2096 - val_accuracy: 0.9838 Epoch 116/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0046 - accuracy: 0.9994 - val_loss: 0.2373 - val_accuracy: 0.9839 Epoch 117/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0028 - accuracy: 0.9995 - val_loss: 0.2777 - val_accuracy: 0.9850 Epoch 118/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0089 - accuracy: 0.9984 - val_loss: 0.3015 - val_accuracy: 0.9821 Epoch 119/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0092 - accuracy: 0.9990 - val_loss: 0.2616 - val_accuracy: 0.9841 Epoch 120/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0056 - accuracy: 0.9990 - val_loss: 0.1801 - val_accuracy: 0.9828 Epoch 121/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0037 - accuracy: 0.9993 - val_loss: 0.3050 - val_accuracy: 0.9812 Epoch 122/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0056 - accuracy: 0.9991 - val_loss: 0.2886 - val_accuracy: 0.9835 Epoch 123/200 469/469 [==============================] - 2s 4ms/step - loss: 7.0380e-04 - accuracy: 0.9998 - val_loss: 0.2756 - val_accuracy: 0.9846 Epoch 124/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0038 - accuracy: 0.9994 - val_loss: 0.2519 - val_accuracy: 0.9832 Epoch 125/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0072 - accuracy: 0.9992 - val_loss: 0.2450 - val_accuracy: 0.9818 Epoch 126/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0075 - accuracy: 0.9987 - val_loss: 0.2015 - val_accuracy: 0.9823 Epoch 127/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0021 - accuracy: 0.9995 - val_loss: 0.2941 - val_accuracy: 0.9796 Epoch 128/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0049 - accuracy: 0.9991 - val_loss: 0.2360 - val_accuracy: 0.9843 Epoch 129/200 469/469 [==============================] - 2s 4ms/step - loss: 7.4236e-04 - accuracy: 0.9999 - val_loss: 0.2844 - val_accuracy: 0.9835 Epoch 130/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0057 - accuracy: 0.9991 - val_loss: 0.1867 - val_accuracy: 0.9806 Epoch 131/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0012 - accuracy: 0.9997 - val_loss: 0.2785 - val_accuracy: 0.9831 Epoch 132/200 469/469 [==============================] - 2s 4ms/step - loss: 3.0945e-04 - accuracy: 0.9999 - val_loss: 0.3273 - val_accuracy: 0.9840 Epoch 133/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9983 - val_loss: 0.1764 - val_accuracy: 0.9827 Epoch 134/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0048 - accuracy: 0.9993 - val_loss: 0.2371 - val_accuracy: 0.9804 Epoch 135/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0031 - accuracy: 0.9996 - val_loss: 0.1933 - val_accuracy: 0.9832 Epoch 136/200 469/469 [==============================] - 2s 3ms/step - loss: 2.0880e-05 - accuracy: 1.0000 - val_loss: 0.2316 - val_accuracy: 0.9829 Epoch 137/200 469/469 [==============================] - 2s 3ms/step - loss: 2.0640e-06 - accuracy: 1.0000 - val_loss: 0.2585 - val_accuracy: 0.9832 Epoch 138/200 469/469 [==============================] - 2s 3ms/step - loss: 7.6387e-07 - accuracy: 1.0000 - val_loss: 0.2776 - val_accuracy: 0.9833 Epoch 139/200 469/469 [==============================] - 2s 3ms/step - loss: 3.1319e-07 - accuracy: 1.0000 - val_loss: 0.2871 - val_accuracy: 0.9832 Epoch 140/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0044e-07 - accuracy: 1.0000 - val_loss: 0.2948 - val_accuracy: 0.9833 Epoch 141/200 469/469 [==============================] - 2s 4ms/step - loss: 1.3919e-07 - accuracy: 1.0000 - val_loss: 0.3027 - val_accuracy: 0.9833 Epoch 142/200 469/469 [==============================] - 2s 3ms/step - loss: 9.6982e-08 - accuracy: 1.0000 - val_loss: 0.3100 - val_accuracy: 0.9832 Epoch 143/200 469/469 [==============================] - 2s 3ms/step - loss: 6.9867e-08 - accuracy: 1.0000 - val_loss: 0.3174 - val_accuracy: 0.9833 Epoch 144/200 469/469 [==============================] - 2s 3ms/step - loss: 5.0588e-08 - accuracy: 1.0000 - val_loss: 0.3242 - val_accuracy: 0.9833 Epoch 145/200 469/469 [==============================] - 2s 3ms/step - loss: 3.7603e-08 - accuracy: 1.0000 - val_loss: 0.3304 - val_accuracy: 0.9833 Epoch 146/200 469/469 [==============================] - 2s 3ms/step - loss: 2.7702e-08 - accuracy: 1.0000 - val_loss: 0.3371 - val_accuracy: 0.9832 Epoch 147/200 469/469 [==============================] - 2s 4ms/step - loss: 2.0692e-08 - accuracy: 1.0000 - val_loss: 0.3431 - val_accuracy: 0.9831 Epoch 148/200 469/469 [==============================] - 2s 4ms/step - loss: 1.5479e-08 - accuracy: 1.0000 - val_loss: 0.3498 - val_accuracy: 0.9831 Epoch 149/200 469/469 [==============================] - 2s 4ms/step - loss: 1.1782e-08 - accuracy: 1.0000 - val_loss: 0.3561 - val_accuracy: 0.9833 Epoch 150/200 469/469 [==============================] - 2s 4ms/step - loss: 9.0956e-09 - accuracy: 1.0000 - val_loss: 0.3621 - val_accuracy: 0.9834 Epoch 151/200 469/469 [==============================] - 2s 3ms/step - loss: 7.1207e-09 - accuracy: 1.0000 - val_loss: 0.3677 - val_accuracy: 0.9834 Epoch 152/200 469/469 [==============================] - 2s 3ms/step - loss: 5.5333e-09 - accuracy: 1.0000 - val_loss: 0.3736 - val_accuracy: 0.9834 Epoch 153/200 469/469 [==============================] - 2s 3ms/step - loss: 4.2776e-09 - accuracy: 1.0000 - val_loss: 0.3791 - val_accuracy: 0.9835 Epoch 154/200 469/469 [==============================] - 2s 3ms/step - loss: 3.3975e-09 - accuracy: 1.0000 - val_loss: 0.3845 - val_accuracy: 0.9835 Epoch 155/200 469/469 [==============================] - 2s 4ms/step - loss: 2.6663e-09 - accuracy: 1.0000 - val_loss: 0.3900 - val_accuracy: 0.9836 Epoch 156/200 469/469 [==============================] - 2s 3ms/step - loss: 2.0941e-09 - accuracy: 1.0000 - val_loss: 0.3957 - val_accuracy: 0.9836 Epoch 157/200 469/469 [==============================] - 2s 3ms/step - loss: 1.6212e-09 - accuracy: 1.0000 - val_loss: 0.4014 - val_accuracy: 0.9836 Epoch 158/200 469/469 [==============================] - 2s 3ms/step - loss: 1.2358e-09 - accuracy: 1.0000 - val_loss: 0.4068 - val_accuracy: 0.9836 Epoch 159/200 469/469 [==============================] - 2s 3ms/step - loss: 9.8745e-10 - accuracy: 1.0000 - val_loss: 0.4117 - val_accuracy: 0.9836 Epoch 160/200 469/469 [==============================] - 2s 3ms/step - loss: 7.7685e-10 - accuracy: 1.0000 - val_loss: 0.4167 - val_accuracy: 0.9837 Epoch 161/200 469/469 [==============================] - 2s 4ms/step - loss: 6.0995e-10 - accuracy: 1.0000 - val_loss: 0.4219 - val_accuracy: 0.9837 Epoch 162/200 469/469 [==============================] - 2s 3ms/step - loss: 4.7684e-10 - accuracy: 1.0000 - val_loss: 0.4268 - val_accuracy: 0.9838 Epoch 163/200 469/469 [==============================] - 2s 3ms/step - loss: 3.7352e-10 - accuracy: 1.0000 - val_loss: 0.4318 - val_accuracy: 0.9839 Epoch 164/200 469/469 [==============================] - 2s 3ms/step - loss: 2.8610e-10 - accuracy: 1.0000 - val_loss: 0.4367 - val_accuracy: 0.9839 Epoch 165/200 469/469 [==============================] - 2s 3ms/step - loss: 2.3047e-10 - accuracy: 1.0000 - val_loss: 0.4410 - val_accuracy: 0.9839 Epoch 166/200 469/469 [==============================] - 2s 3ms/step - loss: 1.8477e-10 - accuracy: 1.0000 - val_loss: 0.4455 - val_accuracy: 0.9839 Epoch 167/200 469/469 [==============================] - 2s 3ms/step - loss: 1.5497e-10 - accuracy: 1.0000 - val_loss: 0.4495 - val_accuracy: 0.9839 Epoch 168/200 469/469 [==============================] - 2s 4ms/step - loss: 1.1921e-10 - accuracy: 1.0000 - val_loss: 0.4536 - val_accuracy: 0.9839 Epoch 169/200 469/469 [==============================] - 2s 3ms/step - loss: 8.9407e-11 - accuracy: 1.0000 - val_loss: 0.4581 - val_accuracy: 0.9839 Epoch 170/200 469/469 [==============================] - 2s 3ms/step - loss: 7.5499e-11 - accuracy: 1.0000 - val_loss: 0.4618 - val_accuracy: 0.9839 Epoch 171/200 469/469 [==============================] - 2s 3ms/step - loss: 6.1591e-11 - accuracy: 1.0000 - val_loss: 0.4654 - val_accuracy: 0.9839 Epoch 172/200 469/469 [==============================] - 2s 3ms/step - loss: 4.7684e-11 - accuracy: 1.0000 - val_loss: 0.4688 - val_accuracy: 0.9839 Epoch 173/200 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-11 - accuracy: 1.0000 - val_loss: 0.4724 - val_accuracy: 0.9839 Epoch 174/200 469/469 [==============================] - 2s 3ms/step - loss: 3.5763e-11 - accuracy: 1.0000 - val_loss: 0.4755 - val_accuracy: 0.9839 Epoch 175/200 469/469 [==============================] - 2s 3ms/step - loss: 2.1855e-11 - accuracy: 1.0000 - val_loss: 0.4791 - val_accuracy: 0.9839 Epoch 176/200 469/469 [==============================] - 2s 4ms/step - loss: 1.3908e-11 - accuracy: 1.0000 - val_loss: 0.4826 - val_accuracy: 0.9840 Epoch 177/200 469/469 [==============================] - 2s 4ms/step - loss: 1.1921e-11 - accuracy: 1.0000 - val_loss: 0.4857 - val_accuracy: 0.9841 Epoch 178/200 469/469 [==============================] - 2s 3ms/step - loss: 1.1921e-11 - accuracy: 1.0000 - val_loss: 0.4885 - val_accuracy: 0.9840 Epoch 179/200 469/469 [==============================] - 2s 4ms/step - loss: 7.9473e-12 - accuracy: 1.0000 - val_loss: 0.4917 - val_accuracy: 0.9841 Epoch 180/200 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.4948 - val_accuracy: 0.9841 Epoch 181/200 469/469 [==============================] - 2s 3ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.4977 - val_accuracy: 0.9842 Epoch 182/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.5008 - val_accuracy: 0.9842 Epoch 183/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.5036 - val_accuracy: 0.9842 Epoch 184/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.5061 - val_accuracy: 0.9842 Epoch 185/200 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.5085 - val_accuracy: 0.9843 Epoch 186/200 469/469 [==============================] - 2s 3ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.5107 - val_accuracy: 0.9843 Epoch 187/200 469/469 [==============================] - 2s 3ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.5128 - val_accuracy: 0.9843 Epoch 188/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5152 - val_accuracy: 0.9843 Epoch 189/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5173 - val_accuracy: 0.9842 Epoch 190/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5195 - val_accuracy: 0.9842 Epoch 191/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5214 - val_accuracy: 0.9842 Epoch 192/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5234 - val_accuracy: 0.9841 Epoch 193/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5252 - val_accuracy: 0.9841 Epoch 194/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5269 - val_accuracy: 0.9841 Epoch 195/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5285 - val_accuracy: 0.9841 Epoch 196/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5301 - val_accuracy: 0.9841 Epoch 197/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5316 - val_accuracy: 0.9841 Epoch 198/200 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5330 - val_accuracy: 0.9841 Epoch 199/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5344 - val_accuracy: 0.9840 Epoch 200/200 469/469 [==============================] - 2s 3ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.5357 - val_accuracy: 0.9840
Visualize Adam Results¶
In [ ]:
#loss_train = history.history['accuracy']
test_acc = historya.history['val_accuracy'][0:18]
test_acc1 = historya1.history['val_accuracy'][0:18]
test_acc2 = historya2.history['val_accuracy'][0:18]
test_acc3 = historya3.history['val_accuracy'][0:18]
test_acc4 = historya4.history['val_accuracy'][0:18]
epochs = range(0,18)
plt.figure(figsize=(20,10))
plt.plot(epochs, test_acc, 'r', label='Logistic: Normal')
plt.plot(epochs, test_acc1, 'r', label='Logistic: Xavier', linewidth=3)
plt.plot(epochs, test_acc2, 'b', label='ReLU: Normal')
plt.plot(epochs, test_acc3, 'b', label='ReLU: Xavier', linewidth=3)
plt.plot(epochs, test_acc4, 'black', label='ReLU: He')
plt.title('Adam')
plt.xlabel('Epoch')
plt.ylabel('Test Accuracy %')
plt.legend()
plt.show()
In [ ]:
#loss_train = history.history['accuracy']
test_acc = historya.history['val_accuracy'][19:]
test_acc1 = historya1.history['val_accuracy'][19:]
test_acc2 = historya2.history['val_accuracy'][19:]
test_acc3 = historya3.history['val_accuracy'][19:]
test_acc4 = historya4.history['val_accuracy'][19:]
epochs = range(19,200)
plt.figure(figsize=(20,10))
plt.plot(epochs, test_acc, 'r', label='Logistic: Normal')
plt.plot(epochs, test_acc1, 'r', label='Logistic: Xavier', linewidth=3)
plt.plot(epochs, test_acc2, 'b', label='ReLU: Normal')
plt.plot(epochs, test_acc3, 'b', label='ReLU: Xavier', linewidth=3)
plt.plot(epochs, test_acc4, 'black', label='ReLU: He')
plt.title('Adam')
plt.xlabel('Epoch')
plt.ylabel('Test Accuracy %')
plt.legend()
plt.show()
Problem 4¶
Activation function: the logistic sigmoid function; initialization: Xavier initializer; no dropout¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.GlorotNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten (Flatten) (None, 784) 0 _________________________________________________________________ dense (Dense) (None, 1024) 803840 _________________________________________________________________ dense_1 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_2 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_3 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_4 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_5 (Dense) (None, 10) 10250 ================================================================= Total params: 5,012,490 Trainable params: 5,012,490 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam()
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"])
history = model.fit(X_train, y_train, batch_size=128, epochs=500, validation_data=(X_test, y_test))
Epoch 1/500 469/469 [==============================] - 2s 5ms/step - loss: 1.0706 - accuracy: 0.5963 - val_loss: 0.3144 - val_accuracy: 0.9059 Epoch 2/500 469/469 [==============================] - 2s 4ms/step - loss: 0.2430 - accuracy: 0.9272 - val_loss: 0.1820 - val_accuracy: 0.9432 Epoch 3/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1600 - accuracy: 0.9518 - val_loss: 0.1574 - val_accuracy: 0.9541 Epoch 4/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1205 - accuracy: 0.9640 - val_loss: 0.1376 - val_accuracy: 0.9572 Epoch 5/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0967 - accuracy: 0.9710 - val_loss: 0.1078 - val_accuracy: 0.9700 Epoch 6/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0774 - accuracy: 0.9765 - val_loss: 0.1397 - val_accuracy: 0.9598 Epoch 7/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0657 - accuracy: 0.9798 - val_loss: 0.1110 - val_accuracy: 0.9693 Epoch 8/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0546 - accuracy: 0.9833 - val_loss: 0.1066 - val_accuracy: 0.9712 Epoch 9/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0469 - accuracy: 0.9857 - val_loss: 0.0909 - val_accuracy: 0.9751 Epoch 10/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0385 - accuracy: 0.9879 - val_loss: 0.0913 - val_accuracy: 0.9732 Epoch 11/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0319 - accuracy: 0.9904 - val_loss: 0.1101 - val_accuracy: 0.9731 Epoch 12/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0303 - accuracy: 0.9905 - val_loss: 0.0909 - val_accuracy: 0.9783 Epoch 13/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0254 - accuracy: 0.9924 - val_loss: 0.1022 - val_accuracy: 0.9748 Epoch 14/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0252 - accuracy: 0.9923 - val_loss: 0.0973 - val_accuracy: 0.9769 Epoch 15/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0201 - accuracy: 0.9942 - val_loss: 0.0888 - val_accuracy: 0.9799 Epoch 16/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0167 - accuracy: 0.9952 - val_loss: 0.0831 - val_accuracy: 0.9812 Epoch 17/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0187 - accuracy: 0.9942 - val_loss: 0.0966 - val_accuracy: 0.9786 Epoch 18/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0155 - accuracy: 0.9954 - val_loss: 0.1140 - val_accuracy: 0.9759 Epoch 19/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0154 - accuracy: 0.9950 - val_loss: 0.1012 - val_accuracy: 0.9798 Epoch 20/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0137 - accuracy: 0.9955 - val_loss: 0.0938 - val_accuracy: 0.9813 Epoch 21/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9963 - val_loss: 0.0929 - val_accuracy: 0.9813 Epoch 22/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0103 - accuracy: 0.9971 - val_loss: 0.0854 - val_accuracy: 0.9824 Epoch 23/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0105 - accuracy: 0.9970 - val_loss: 0.0861 - val_accuracy: 0.9829 Epoch 24/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0098 - accuracy: 0.9971 - val_loss: 0.1058 - val_accuracy: 0.9809 Epoch 25/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9966 - val_loss: 0.0934 - val_accuracy: 0.9832 Epoch 26/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0074 - accuracy: 0.9979 - val_loss: 0.1188 - val_accuracy: 0.9814 Epoch 27/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9972 - val_loss: 0.1010 - val_accuracy: 0.9815 Epoch 28/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0060 - accuracy: 0.9981 - val_loss: 0.0933 - val_accuracy: 0.9830 Epoch 29/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0088 - accuracy: 0.9974 - val_loss: 0.1032 - val_accuracy: 0.9815 Epoch 30/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0046 - accuracy: 0.9987 - val_loss: 0.1225 - val_accuracy: 0.9815 Epoch 31/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0084 - accuracy: 0.9977 - val_loss: 0.1092 - val_accuracy: 0.9820 Epoch 32/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0059 - accuracy: 0.9986 - val_loss: 0.1302 - val_accuracy: 0.9746 Epoch 33/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0069 - accuracy: 0.9981 - val_loss: 0.1088 - val_accuracy: 0.9834 Epoch 34/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0061 - accuracy: 0.9984 - val_loss: 0.1233 - val_accuracy: 0.9802 Epoch 35/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0076 - accuracy: 0.9979 - val_loss: 0.0882 - val_accuracy: 0.9826 Epoch 36/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0053 - accuracy: 0.9984 - val_loss: 0.0975 - val_accuracy: 0.9831 Epoch 37/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0042 - accuracy: 0.9988 - val_loss: 0.1068 - val_accuracy: 0.9820 Epoch 38/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0058 - accuracy: 0.9984 - val_loss: 0.1038 - val_accuracy: 0.9817 Epoch 39/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0043 - accuracy: 0.9987 - val_loss: 0.1000 - val_accuracy: 0.9835 Epoch 40/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0050 - accuracy: 0.9987 - val_loss: 0.0965 - val_accuracy: 0.9850 Epoch 41/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0049 - accuracy: 0.9985 - val_loss: 0.0951 - val_accuracy: 0.9822 Epoch 42/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0051 - accuracy: 0.9986 - val_loss: 0.1026 - val_accuracy: 0.9850 Epoch 43/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9988 - val_loss: 0.1345 - val_accuracy: 0.9779 Epoch 44/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0040 - accuracy: 0.9990 - val_loss: 0.1131 - val_accuracy: 0.9836 Epoch 45/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0069 - accuracy: 0.9982 - val_loss: 0.1085 - val_accuracy: 0.9832 Epoch 46/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0041 - accuracy: 0.9988 - val_loss: 0.1179 - val_accuracy: 0.9829 Epoch 47/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0044 - accuracy: 0.9988 - val_loss: 0.1326 - val_accuracy: 0.9783 Epoch 48/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0043 - accuracy: 0.9987 - val_loss: 0.0946 - val_accuracy: 0.9849 Epoch 49/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9990 - val_loss: 0.1099 - val_accuracy: 0.9806 Epoch 50/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0044 - accuracy: 0.9988 - val_loss: 0.1021 - val_accuracy: 0.9839 Epoch 51/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0029 - accuracy: 0.9993 - val_loss: 0.1058 - val_accuracy: 0.9843 Epoch 52/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0023 - accuracy: 0.9994 - val_loss: 0.1125 - val_accuracy: 0.9815 Epoch 53/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0031 - accuracy: 0.9991 - val_loss: 0.1121 - val_accuracy: 0.9813 Epoch 54/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0031 - accuracy: 0.9991 - val_loss: 0.1285 - val_accuracy: 0.9819 Epoch 55/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0051 - accuracy: 0.9984 - val_loss: 0.1205 - val_accuracy: 0.9814 Epoch 56/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0039 - accuracy: 0.9990 - val_loss: 0.1145 - val_accuracy: 0.9826 Epoch 57/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0029 - accuracy: 0.9993 - val_loss: 0.1089 - val_accuracy: 0.9829 Epoch 58/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9990 - val_loss: 0.1310 - val_accuracy: 0.9818 Epoch 59/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9990 - val_loss: 0.1104 - val_accuracy: 0.9827 Epoch 60/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0019 - accuracy: 0.9995 - val_loss: 0.1262 - val_accuracy: 0.9820 Epoch 61/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0029 - accuracy: 0.9993 - val_loss: 0.1172 - val_accuracy: 0.9828 Epoch 62/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0042 - accuracy: 0.9990 - val_loss: 0.1235 - val_accuracy: 0.9813 Epoch 63/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0030 - accuracy: 0.9991 - val_loss: 0.1069 - val_accuracy: 0.9839 Epoch 64/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0033 - accuracy: 0.9992 - val_loss: 0.1160 - val_accuracy: 0.9822 Epoch 65/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0022 - accuracy: 0.9994 - val_loss: 0.1080 - val_accuracy: 0.9840 Epoch 66/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0020 - accuracy: 0.9994 - val_loss: 0.1197 - val_accuracy: 0.9829 Epoch 67/500 469/469 [==============================] - 2s 4ms/step - loss: 7.0570e-04 - accuracy: 0.9998 - val_loss: 0.1256 - val_accuracy: 0.9845 Epoch 68/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0045 - accuracy: 0.9985 - val_loss: 0.1114 - val_accuracy: 0.9839 Epoch 69/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9990 - val_loss: 0.1215 - val_accuracy: 0.9811 Epoch 70/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0024 - accuracy: 0.9994 - val_loss: 0.1130 - val_accuracy: 0.9832 Epoch 71/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0012 - accuracy: 0.9996 - val_loss: 0.1248 - val_accuracy: 0.9831 Epoch 72/500 469/469 [==============================] - 2s 4ms/step - loss: 8.6865e-04 - accuracy: 0.9998 - val_loss: 0.1292 - val_accuracy: 0.9833 Epoch 73/500 469/469 [==============================] - 2s 4ms/step - loss: 3.3745e-04 - accuracy: 0.9999 - val_loss: 0.1524 - val_accuracy: 0.9828 Epoch 74/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0063 - accuracy: 0.9984 - val_loss: 0.1209 - val_accuracy: 0.9816 Epoch 75/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0030 - accuracy: 0.9992 - val_loss: 0.1049 - val_accuracy: 0.9837 Epoch 76/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0013 - accuracy: 0.9995 - val_loss: 0.1192 - val_accuracy: 0.9842 Epoch 77/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0033 - accuracy: 0.9992 - val_loss: 0.1078 - val_accuracy: 0.9834 Epoch 78/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0023 - accuracy: 0.9994 - val_loss: 0.1049 - val_accuracy: 0.9845 Epoch 79/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0014 - accuracy: 0.9996 - val_loss: 0.1463 - val_accuracy: 0.9806 Epoch 80/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0020 - accuracy: 0.9996 - val_loss: 0.1152 - val_accuracy: 0.9827 Epoch 81/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0038 - accuracy: 0.9989 - val_loss: 0.1263 - val_accuracy: 0.9816 Epoch 82/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0016 - accuracy: 0.9995 - val_loss: 0.1276 - val_accuracy: 0.9814 Epoch 83/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0010 - accuracy: 0.9998 - val_loss: 0.1174 - val_accuracy: 0.9835 Epoch 84/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0010 - accuracy: 0.9998 - val_loss: 0.1194 - val_accuracy: 0.9838 Epoch 85/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0017 - accuracy: 0.9995 - val_loss: 0.1220 - val_accuracy: 0.9844 Epoch 86/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0015 - accuracy: 0.9997 - val_loss: 0.1310 - val_accuracy: 0.9828 Epoch 87/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0054 - accuracy: 0.9987 - val_loss: 0.1514 - val_accuracy: 0.9784 Epoch 88/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0035 - accuracy: 0.9991 - val_loss: 0.1030 - val_accuracy: 0.9840 Epoch 89/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0016 - accuracy: 0.9995 - val_loss: 0.1090 - val_accuracy: 0.9841 Epoch 90/500 469/469 [==============================] - 2s 4ms/step - loss: 8.4610e-04 - accuracy: 0.9998 - val_loss: 0.1194 - val_accuracy: 0.9837 Epoch 91/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0022 - accuracy: 0.9993 - val_loss: 0.1230 - val_accuracy: 0.9844 Epoch 92/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0013 - accuracy: 0.9996 - val_loss: 0.1356 - val_accuracy: 0.9839 Epoch 93/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0043 - accuracy: 0.9990 - val_loss: 0.1076 - val_accuracy: 0.9821 Epoch 94/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0031 - accuracy: 0.9992 - val_loss: 0.1012 - val_accuracy: 0.9852 Epoch 95/500 469/469 [==============================] - 2s 4ms/step - loss: 8.1479e-04 - accuracy: 0.9998 - val_loss: 0.1112 - val_accuracy: 0.9856 Epoch 96/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5384e-04 - accuracy: 0.9999 - val_loss: 0.1177 - val_accuracy: 0.9857 Epoch 97/500 469/469 [==============================] - 2s 4ms/step - loss: 9.3516e-06 - accuracy: 1.0000 - val_loss: 0.1214 - val_accuracy: 0.9857 Epoch 98/500 469/469 [==============================] - 2s 4ms/step - loss: 4.5289e-06 - accuracy: 1.0000 - val_loss: 0.1248 - val_accuracy: 0.9857 Epoch 99/500 469/469 [==============================] - 2s 4ms/step - loss: 3.0145e-06 - accuracy: 1.0000 - val_loss: 0.1279 - val_accuracy: 0.9856 Epoch 100/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1782e-06 - accuracy: 1.0000 - val_loss: 0.1307 - val_accuracy: 0.9856 Epoch 101/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6233e-06 - accuracy: 1.0000 - val_loss: 0.1334 - val_accuracy: 0.9856 Epoch 102/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2253e-06 - accuracy: 1.0000 - val_loss: 0.1359 - val_accuracy: 0.9856 Epoch 103/500 469/469 [==============================] - 2s 4ms/step - loss: 9.3665e-07 - accuracy: 1.0000 - val_loss: 0.1382 - val_accuracy: 0.9855 Epoch 104/500 469/469 [==============================] - 2s 4ms/step - loss: 7.3005e-07 - accuracy: 1.0000 - val_loss: 0.1405 - val_accuracy: 0.9854 Epoch 105/500 469/469 [==============================] - 2s 4ms/step - loss: 5.6435e-07 - accuracy: 1.0000 - val_loss: 0.1428 - val_accuracy: 0.9854 Epoch 106/500 469/469 [==============================] - 2s 4ms/step - loss: 4.4435e-07 - accuracy: 1.0000 - val_loss: 0.1449 - val_accuracy: 0.9854 Epoch 107/500 469/469 [==============================] - 2s 4ms/step - loss: 3.5198e-07 - accuracy: 1.0000 - val_loss: 0.1470 - val_accuracy: 0.9854 Epoch 108/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7838e-07 - accuracy: 1.0000 - val_loss: 0.1490 - val_accuracy: 0.9854 Epoch 109/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2501e-07 - accuracy: 1.0000 - val_loss: 0.1509 - val_accuracy: 0.9854 Epoch 110/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8387e-07 - accuracy: 1.0000 - val_loss: 0.1527 - val_accuracy: 0.9854 Epoch 111/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5080e-07 - accuracy: 1.0000 - val_loss: 0.1545 - val_accuracy: 0.9855 Epoch 112/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2090e-07 - accuracy: 1.0000 - val_loss: 0.1563 - val_accuracy: 0.9855 Epoch 113/500 469/469 [==============================] - 2s 4ms/step - loss: 9.4881e-08 - accuracy: 1.0000 - val_loss: 0.1579 - val_accuracy: 0.9855 Epoch 114/500 469/469 [==============================] - 2s 4ms/step - loss: 7.6890e-08 - accuracy: 1.0000 - val_loss: 0.1595 - val_accuracy: 0.9856 Epoch 115/500 469/469 [==============================] - 2s 4ms/step - loss: 6.3334e-08 - accuracy: 1.0000 - val_loss: 0.1610 - val_accuracy: 0.9857 Epoch 116/500 469/469 [==============================] - 2s 4ms/step - loss: 5.2881e-08 - accuracy: 1.0000 - val_loss: 0.1625 - val_accuracy: 0.9857 Epoch 117/500 469/469 [==============================] - 2s 4ms/step - loss: 4.4438e-08 - accuracy: 1.0000 - val_loss: 0.1639 - val_accuracy: 0.9857 Epoch 118/500 469/469 [==============================] - 2s 4ms/step - loss: 3.7847e-08 - accuracy: 1.0000 - val_loss: 0.1652 - val_accuracy: 0.9857 Epoch 119/500 469/469 [==============================] - 2s 4ms/step - loss: 3.2166e-08 - accuracy: 1.0000 - val_loss: 0.1665 - val_accuracy: 0.9857 Epoch 120/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7555e-08 - accuracy: 1.0000 - val_loss: 0.1677 - val_accuracy: 0.9857 Epoch 121/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3654e-08 - accuracy: 1.0000 - val_loss: 0.1689 - val_accuracy: 0.9857 Epoch 122/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0280e-08 - accuracy: 1.0000 - val_loss: 0.1700 - val_accuracy: 0.9857 Epoch 123/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7577e-08 - accuracy: 1.0000 - val_loss: 0.1711 - val_accuracy: 0.9857 Epoch 124/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5213e-08 - accuracy: 1.0000 - val_loss: 0.1721 - val_accuracy: 0.9858 Epoch 125/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3274e-08 - accuracy: 1.0000 - val_loss: 0.1732 - val_accuracy: 0.9858 Epoch 126/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1646e-08 - accuracy: 1.0000 - val_loss: 0.1741 - val_accuracy: 0.9858 Epoch 127/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0288e-08 - accuracy: 1.0000 - val_loss: 0.1750 - val_accuracy: 0.9858 Epoch 128/500 469/469 [==============================] - 2s 4ms/step - loss: 9.1193e-09 - accuracy: 1.0000 - val_loss: 0.1759 - val_accuracy: 0.9858 Epoch 129/500 469/469 [==============================] - 2s 4ms/step - loss: 8.1141e-09 - accuracy: 1.0000 - val_loss: 0.1767 - val_accuracy: 0.9858 Epoch 130/500 469/469 [==============================] - 2s 4ms/step - loss: 7.2995e-09 - accuracy: 1.0000 - val_loss: 0.1775 - val_accuracy: 0.9859 Epoch 131/500 469/469 [==============================] - 2s 4ms/step - loss: 6.5405e-09 - accuracy: 1.0000 - val_loss: 0.1783 - val_accuracy: 0.9859 Epoch 132/500 469/469 [==============================] - 2s 4ms/step - loss: 5.9048e-09 - accuracy: 1.0000 - val_loss: 0.1790 - val_accuracy: 0.9860 Epoch 133/500 469/469 [==============================] - 2s 4ms/step - loss: 5.3723e-09 - accuracy: 1.0000 - val_loss: 0.1797 - val_accuracy: 0.9860 Epoch 134/500 469/469 [==============================] - 2s 4ms/step - loss: 4.8875e-09 - accuracy: 1.0000 - val_loss: 0.1804 - val_accuracy: 0.9860 Epoch 135/500 469/469 [==============================] - 2s 4ms/step - loss: 4.4485e-09 - accuracy: 1.0000 - val_loss: 0.1810 - val_accuracy: 0.9859 Epoch 136/500 469/469 [==============================] - 2s 4ms/step - loss: 4.1326e-09 - accuracy: 1.0000 - val_loss: 0.1816 - val_accuracy: 0.9859 Epoch 137/500 469/469 [==============================] - 2s 4ms/step - loss: 3.8286e-09 - accuracy: 1.0000 - val_loss: 0.1822 - val_accuracy: 0.9859 Epoch 138/500 469/469 [==============================] - 2s 4ms/step - loss: 3.5266e-09 - accuracy: 1.0000 - val_loss: 0.1827 - val_accuracy: 0.9859 Epoch 139/500 469/469 [==============================] - 2s 4ms/step - loss: 3.2822e-09 - accuracy: 1.0000 - val_loss: 0.1833 - val_accuracy: 0.9858 Epoch 140/500 469/469 [==============================] - 2s 4ms/step - loss: 3.0637e-09 - accuracy: 1.0000 - val_loss: 0.1838 - val_accuracy: 0.9858 Epoch 141/500 469/469 [==============================] - 2s 4ms/step - loss: 2.8749e-09 - accuracy: 1.0000 - val_loss: 0.1842 - val_accuracy: 0.9858 Epoch 142/500 469/469 [==============================] - 2s 4ms/step - loss: 2.6961e-09 - accuracy: 1.0000 - val_loss: 0.1846 - val_accuracy: 0.9858 Epoch 143/500 469/469 [==============================] - 2s 4ms/step - loss: 2.5471e-09 - accuracy: 1.0000 - val_loss: 0.1850 - val_accuracy: 0.9858 Epoch 144/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4219e-09 - accuracy: 1.0000 - val_loss: 0.1855 - val_accuracy: 0.9858 Epoch 145/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2987e-09 - accuracy: 1.0000 - val_loss: 0.1858 - val_accuracy: 0.9859 Epoch 146/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1895e-09 - accuracy: 1.0000 - val_loss: 0.1862 - val_accuracy: 0.9859 Epoch 147/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0742e-09 - accuracy: 1.0000 - val_loss: 0.1866 - val_accuracy: 0.9859 Epoch 148/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9749e-09 - accuracy: 1.0000 - val_loss: 0.1869 - val_accuracy: 0.9859 Epoch 149/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9014e-09 - accuracy: 1.0000 - val_loss: 0.1872 - val_accuracy: 0.9859 Epoch 150/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8159e-09 - accuracy: 1.0000 - val_loss: 0.1875 - val_accuracy: 0.9859 Epoch 151/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7484e-09 - accuracy: 1.0000 - val_loss: 0.1878 - val_accuracy: 0.9859 Epoch 152/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6908e-09 - accuracy: 1.0000 - val_loss: 0.1881 - val_accuracy: 0.9859 Epoch 153/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6312e-09 - accuracy: 1.0000 - val_loss: 0.1884 - val_accuracy: 0.9859 Epoch 154/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5557e-09 - accuracy: 1.0000 - val_loss: 0.1887 - val_accuracy: 0.9859 Epoch 155/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5120e-09 - accuracy: 1.0000 - val_loss: 0.1889 - val_accuracy: 0.9859 Epoch 156/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4583e-09 - accuracy: 1.0000 - val_loss: 0.1892 - val_accuracy: 0.9859 Epoch 157/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4206e-09 - accuracy: 1.0000 - val_loss: 0.1894 - val_accuracy: 0.9859 Epoch 158/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3689e-09 - accuracy: 1.0000 - val_loss: 0.1897 - val_accuracy: 0.9859 Epoch 159/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3252e-09 - accuracy: 1.0000 - val_loss: 0.1899 - val_accuracy: 0.9859 Epoch 160/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2855e-09 - accuracy: 1.0000 - val_loss: 0.1901 - val_accuracy: 0.9859 Epoch 161/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2457e-09 - accuracy: 1.0000 - val_loss: 0.1904 - val_accuracy: 0.9859 Epoch 162/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1981e-09 - accuracy: 1.0000 - val_loss: 0.1906 - val_accuracy: 0.9859 Epoch 163/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1802e-09 - accuracy: 1.0000 - val_loss: 0.1908 - val_accuracy: 0.9859 Epoch 164/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1404e-09 - accuracy: 1.0000 - val_loss: 0.1910 - val_accuracy: 0.9859 Epoch 165/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1265e-09 - accuracy: 1.0000 - val_loss: 0.1912 - val_accuracy: 0.9859 Epoch 166/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0808e-09 - accuracy: 1.0000 - val_loss: 0.1914 - val_accuracy: 0.9859 Epoch 167/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0510e-09 - accuracy: 1.0000 - val_loss: 0.1915 - val_accuracy: 0.9859 Epoch 168/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0312e-09 - accuracy: 1.0000 - val_loss: 0.1917 - val_accuracy: 0.9859 Epoch 169/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0014e-09 - accuracy: 1.0000 - val_loss: 0.1919 - val_accuracy: 0.9859 Epoch 170/500 469/469 [==============================] - 2s 4ms/step - loss: 9.7751e-10 - accuracy: 1.0000 - val_loss: 0.1921 - val_accuracy: 0.9859 Epoch 171/500 469/469 [==============================] - 2s 4ms/step - loss: 9.4970e-10 - accuracy: 1.0000 - val_loss: 0.1922 - val_accuracy: 0.9859 Epoch 172/500 469/469 [==============================] - 2s 4ms/step - loss: 9.2983e-10 - accuracy: 1.0000 - val_loss: 0.1924 - val_accuracy: 0.9859 Epoch 173/500 469/469 [==============================] - 2s 4ms/step - loss: 9.0400e-10 - accuracy: 1.0000 - val_loss: 0.1926 - val_accuracy: 0.9859 Epoch 174/500 469/469 [==============================] - 2s 4ms/step - loss: 8.8215e-10 - accuracy: 1.0000 - val_loss: 0.1927 - val_accuracy: 0.9859 Epoch 175/500 469/469 [==============================] - 2s 4ms/step - loss: 8.6228e-10 - accuracy: 1.0000 - val_loss: 0.1929 - val_accuracy: 0.9859 Epoch 176/500 469/469 [==============================] - 2s 4ms/step - loss: 8.4440e-10 - accuracy: 1.0000 - val_loss: 0.1930 - val_accuracy: 0.9860 Epoch 177/500 469/469 [==============================] - 2s 4ms/step - loss: 8.2850e-10 - accuracy: 1.0000 - val_loss: 0.1932 - val_accuracy: 0.9860 Epoch 178/500 469/469 [==============================] - 2s 4ms/step - loss: 8.0665e-10 - accuracy: 1.0000 - val_loss: 0.1933 - val_accuracy: 0.9860 Epoch 179/500 469/469 [==============================] - 2s 4ms/step - loss: 7.8281e-10 - accuracy: 1.0000 - val_loss: 0.1934 - val_accuracy: 0.9860 Epoch 180/500 469/469 [==============================] - 2s 4ms/step - loss: 7.7287e-10 - accuracy: 1.0000 - val_loss: 0.1936 - val_accuracy: 0.9860 Epoch 181/500 469/469 [==============================] - 2s 4ms/step - loss: 7.6095e-10 - accuracy: 1.0000 - val_loss: 0.1937 - val_accuracy: 0.9860 Epoch 182/500 469/469 [==============================] - 2s 4ms/step - loss: 7.4704e-10 - accuracy: 1.0000 - val_loss: 0.1939 - val_accuracy: 0.9860 Epoch 183/500 469/469 [==============================] - 2s 4ms/step - loss: 7.3314e-10 - accuracy: 1.0000 - val_loss: 0.1940 - val_accuracy: 0.9860 Epoch 184/500 469/469 [==============================] - 2s 4ms/step - loss: 7.1327e-10 - accuracy: 1.0000 - val_loss: 0.1941 - val_accuracy: 0.9860 Epoch 185/500 469/469 [==============================] - 2s 4ms/step - loss: 7.0333e-10 - accuracy: 1.0000 - val_loss: 0.1942 - val_accuracy: 0.9860 Epoch 186/500 469/469 [==============================] - 2s 4ms/step - loss: 6.8545e-10 - accuracy: 1.0000 - val_loss: 0.1944 - val_accuracy: 0.9860 Epoch 187/500 469/469 [==============================] - 2s 4ms/step - loss: 6.7353e-10 - accuracy: 1.0000 - val_loss: 0.1945 - val_accuracy: 0.9860 Epoch 188/500 469/469 [==============================] - 2s 4ms/step - loss: 6.6161e-10 - accuracy: 1.0000 - val_loss: 0.1946 - val_accuracy: 0.9860 Epoch 189/500 469/469 [==============================] - 2s 4ms/step - loss: 6.4770e-10 - accuracy: 1.0000 - val_loss: 0.1947 - val_accuracy: 0.9861 Epoch 190/500 469/469 [==============================] - 2s 5ms/step - loss: 6.3380e-10 - accuracy: 1.0000 - val_loss: 0.1948 - val_accuracy: 0.9861 Epoch 191/500 469/469 [==============================] - 2s 5ms/step - loss: 6.1989e-10 - accuracy: 1.0000 - val_loss: 0.1949 - val_accuracy: 0.9861 Epoch 192/500 469/469 [==============================] - 2s 5ms/step - loss: 6.0995e-10 - accuracy: 1.0000 - val_loss: 0.1950 - val_accuracy: 0.9861 Epoch 193/500 469/469 [==============================] - 2s 4ms/step - loss: 6.0399e-10 - accuracy: 1.0000 - val_loss: 0.1952 - val_accuracy: 0.9861 Epoch 194/500 469/469 [==============================] - 2s 4ms/step - loss: 6.0002e-10 - accuracy: 1.0000 - val_loss: 0.1953 - val_accuracy: 0.9861 Epoch 195/500 469/469 [==============================] - 2s 4ms/step - loss: 5.9009e-10 - accuracy: 1.0000 - val_loss: 0.1954 - val_accuracy: 0.9861 Epoch 196/500 469/469 [==============================] - 2s 4ms/step - loss: 5.8611e-10 - accuracy: 1.0000 - val_loss: 0.1955 - val_accuracy: 0.9861 Epoch 197/500 469/469 [==============================] - 2s 4ms/step - loss: 5.8412e-10 - accuracy: 1.0000 - val_loss: 0.1956 - val_accuracy: 0.9861 Epoch 198/500 469/469 [==============================] - 2s 4ms/step - loss: 5.7419e-10 - accuracy: 1.0000 - val_loss: 0.1957 - val_accuracy: 0.9861 Epoch 199/500 469/469 [==============================] - 2s 4ms/step - loss: 5.6227e-10 - accuracy: 1.0000 - val_loss: 0.1958 - val_accuracy: 0.9861 Epoch 200/500 469/469 [==============================] - 2s 4ms/step - loss: 5.5432e-10 - accuracy: 1.0000 - val_loss: 0.1959 - val_accuracy: 0.9861 Epoch 201/500 469/469 [==============================] - 2s 4ms/step - loss: 5.4439e-10 - accuracy: 1.0000 - val_loss: 0.1960 - val_accuracy: 0.9861 Epoch 202/500 469/469 [==============================] - 2s 4ms/step - loss: 5.3445e-10 - accuracy: 1.0000 - val_loss: 0.1961 - val_accuracy: 0.9861 Epoch 203/500 469/469 [==============================] - 2s 4ms/step - loss: 5.3048e-10 - accuracy: 1.0000 - val_loss: 0.1962 - val_accuracy: 0.9861 Epoch 204/500 469/469 [==============================] - 2s 4ms/step - loss: 5.2452e-10 - accuracy: 1.0000 - val_loss: 0.1963 - val_accuracy: 0.9861 Epoch 205/500 469/469 [==============================] - 2s 4ms/step - loss: 5.1260e-10 - accuracy: 1.0000 - val_loss: 0.1964 - val_accuracy: 0.9861 Epoch 206/500 469/469 [==============================] - 2s 4ms/step - loss: 5.0267e-10 - accuracy: 1.0000 - val_loss: 0.1965 - val_accuracy: 0.9861 Epoch 207/500 469/469 [==============================] - 2s 4ms/step - loss: 4.9670e-10 - accuracy: 1.0000 - val_loss: 0.1966 - val_accuracy: 0.9861 Epoch 208/500 469/469 [==============================] - 2s 4ms/step - loss: 4.9074e-10 - accuracy: 1.0000 - val_loss: 0.1966 - val_accuracy: 0.9861 Epoch 209/500 469/469 [==============================] - 2s 4ms/step - loss: 4.8280e-10 - accuracy: 1.0000 - val_loss: 0.1967 - val_accuracy: 0.9861 Epoch 210/500 469/469 [==============================] - 2s 4ms/step - loss: 4.7882e-10 - accuracy: 1.0000 - val_loss: 0.1968 - val_accuracy: 0.9861 Epoch 211/500 469/469 [==============================] - 2s 4ms/step - loss: 4.7684e-10 - accuracy: 1.0000 - val_loss: 0.1969 - val_accuracy: 0.9861 Epoch 212/500 469/469 [==============================] - 2s 4ms/step - loss: 4.6889e-10 - accuracy: 1.0000 - val_loss: 0.1970 - val_accuracy: 0.9861 Epoch 213/500 469/469 [==============================] - 2s 4ms/step - loss: 4.6492e-10 - accuracy: 1.0000 - val_loss: 0.1971 - val_accuracy: 0.9861 Epoch 214/500 469/469 [==============================] - 2s 4ms/step - loss: 4.5896e-10 - accuracy: 1.0000 - val_loss: 0.1972 - val_accuracy: 0.9861 Epoch 215/500 469/469 [==============================] - 2s 4ms/step - loss: 4.5498e-10 - accuracy: 1.0000 - val_loss: 0.1972 - val_accuracy: 0.9861 Epoch 216/500 469/469 [==============================] - 2s 4ms/step - loss: 4.4902e-10 - accuracy: 1.0000 - val_loss: 0.1973 - val_accuracy: 0.9861 Epoch 217/500 469/469 [==============================] - 2s 4ms/step - loss: 4.4703e-10 - accuracy: 1.0000 - val_loss: 0.1974 - val_accuracy: 0.9861 Epoch 218/500 469/469 [==============================] - 2s 4ms/step - loss: 4.4306e-10 - accuracy: 1.0000 - val_loss: 0.1975 - val_accuracy: 0.9861 Epoch 219/500 469/469 [==============================] - 2s 4ms/step - loss: 4.3511e-10 - accuracy: 1.0000 - val_loss: 0.1976 - val_accuracy: 0.9861 Epoch 220/500 469/469 [==============================] - 2s 4ms/step - loss: 4.2717e-10 - accuracy: 1.0000 - val_loss: 0.1977 - val_accuracy: 0.9861 Epoch 221/500 469/469 [==============================] - 2s 4ms/step - loss: 4.2121e-10 - accuracy: 1.0000 - val_loss: 0.1977 - val_accuracy: 0.9862 Epoch 222/500 469/469 [==============================] - 2s 4ms/step - loss: 4.2121e-10 - accuracy: 1.0000 - val_loss: 0.1978 - val_accuracy: 0.9862 Epoch 223/500 469/469 [==============================] - 2s 4ms/step - loss: 4.1127e-10 - accuracy: 1.0000 - val_loss: 0.1979 - val_accuracy: 0.9862 Epoch 224/500 469/469 [==============================] - 2s 4ms/step - loss: 4.0730e-10 - accuracy: 1.0000 - val_loss: 0.1980 - val_accuracy: 0.9862 Epoch 225/500 469/469 [==============================] - 2s 4ms/step - loss: 4.0531e-10 - accuracy: 1.0000 - val_loss: 0.1980 - val_accuracy: 0.9862 Epoch 226/500 469/469 [==============================] - 2s 4ms/step - loss: 4.0134e-10 - accuracy: 1.0000 - val_loss: 0.1981 - val_accuracy: 0.9862 Epoch 227/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9538e-10 - accuracy: 1.0000 - val_loss: 0.1982 - val_accuracy: 0.9862 Epoch 228/500 469/469 [==============================] - 2s 4ms/step - loss: 3.8942e-10 - accuracy: 1.0000 - val_loss: 0.1982 - val_accuracy: 0.9862 Epoch 229/500 469/469 [==============================] - 2s 4ms/step - loss: 3.8346e-10 - accuracy: 1.0000 - val_loss: 0.1983 - val_accuracy: 0.9862 Epoch 230/500 469/469 [==============================] - 2s 4ms/step - loss: 3.8346e-10 - accuracy: 1.0000 - val_loss: 0.1984 - val_accuracy: 0.9862 Epoch 231/500 469/469 [==============================] - 2s 4ms/step - loss: 3.7750e-10 - accuracy: 1.0000 - val_loss: 0.1985 - val_accuracy: 0.9862 Epoch 232/500 469/469 [==============================] - 2s 4ms/step - loss: 3.7551e-10 - accuracy: 1.0000 - val_loss: 0.1985 - val_accuracy: 0.9862 Epoch 233/500 469/469 [==============================] - 2s 4ms/step - loss: 3.7352e-10 - accuracy: 1.0000 - val_loss: 0.1986 - val_accuracy: 0.9862 Epoch 234/500 469/469 [==============================] - 2s 4ms/step - loss: 3.6557e-10 - accuracy: 1.0000 - val_loss: 0.1987 - val_accuracy: 0.9862 Epoch 235/500 469/469 [==============================] - 2s 4ms/step - loss: 3.6359e-10 - accuracy: 1.0000 - val_loss: 0.1987 - val_accuracy: 0.9862 Epoch 236/500 469/469 [==============================] - 2s 4ms/step - loss: 3.5763e-10 - accuracy: 1.0000 - val_loss: 0.1988 - val_accuracy: 0.9862 Epoch 237/500 469/469 [==============================] - 2s 4ms/step - loss: 3.5365e-10 - accuracy: 1.0000 - val_loss: 0.1989 - val_accuracy: 0.9862 Epoch 238/500 469/469 [==============================] - 2s 4ms/step - loss: 3.4968e-10 - accuracy: 1.0000 - val_loss: 0.1989 - val_accuracy: 0.9862 Epoch 239/500 469/469 [==============================] - 2s 4ms/step - loss: 3.4968e-10 - accuracy: 1.0000 - val_loss: 0.1990 - val_accuracy: 0.9862 Epoch 240/500 469/469 [==============================] - 2s 4ms/step - loss: 3.4372e-10 - accuracy: 1.0000 - val_loss: 0.1991 - val_accuracy: 0.9862 Epoch 241/500 469/469 [==============================] - 2s 4ms/step - loss: 3.4173e-10 - accuracy: 1.0000 - val_loss: 0.1991 - val_accuracy: 0.9862 Epoch 242/500 469/469 [==============================] - 2s 4ms/step - loss: 3.4173e-10 - accuracy: 1.0000 - val_loss: 0.1992 - val_accuracy: 0.9862 Epoch 243/500 469/469 [==============================] - 2s 4ms/step - loss: 3.3776e-10 - accuracy: 1.0000 - val_loss: 0.1992 - val_accuracy: 0.9862 Epoch 244/500 469/469 [==============================] - 2s 4ms/step - loss: 3.3577e-10 - accuracy: 1.0000 - val_loss: 0.1993 - val_accuracy: 0.9862 Epoch 245/500 469/469 [==============================] - 2s 4ms/step - loss: 3.3379e-10 - accuracy: 1.0000 - val_loss: 0.1994 - val_accuracy: 0.9862 Epoch 246/500 469/469 [==============================] - 2s 4ms/step - loss: 3.2981e-10 - accuracy: 1.0000 - val_loss: 0.1994 - val_accuracy: 0.9862 Epoch 247/500 469/469 [==============================] - 2s 4ms/step - loss: 3.2783e-10 - accuracy: 1.0000 - val_loss: 0.1995 - val_accuracy: 0.9862 Epoch 248/500 469/469 [==============================] - 2s 4ms/step - loss: 3.2385e-10 - accuracy: 1.0000 - val_loss: 0.1996 - val_accuracy: 0.9862 Epoch 249/500 469/469 [==============================] - 2s 4ms/step - loss: 3.2186e-10 - accuracy: 1.0000 - val_loss: 0.1996 - val_accuracy: 0.9862 Epoch 250/500 469/469 [==============================] - 2s 4ms/step - loss: 3.1988e-10 - accuracy: 1.0000 - val_loss: 0.1997 - val_accuracy: 0.9862 Epoch 251/500 469/469 [==============================] - 2s 4ms/step - loss: 3.1988e-10 - accuracy: 1.0000 - val_loss: 0.1997 - val_accuracy: 0.9862 Epoch 252/500 469/469 [==============================] - 2s 4ms/step - loss: 3.1988e-10 - accuracy: 1.0000 - val_loss: 0.1998 - val_accuracy: 0.9862 Epoch 253/500 469/469 [==============================] - 2s 4ms/step - loss: 3.1392e-10 - accuracy: 1.0000 - val_loss: 0.1998 - val_accuracy: 0.9862 Epoch 254/500 469/469 [==============================] - 2s 4ms/step - loss: 3.0994e-10 - accuracy: 1.0000 - val_loss: 0.1999 - val_accuracy: 0.9862 Epoch 255/500 469/469 [==============================] - 2s 4ms/step - loss: 3.0994e-10 - accuracy: 1.0000 - val_loss: 0.2000 - val_accuracy: 0.9862 Epoch 256/500 469/469 [==============================] - 2s 4ms/step - loss: 3.0398e-10 - accuracy: 1.0000 - val_loss: 0.2000 - val_accuracy: 0.9862 Epoch 257/500 469/469 [==============================] - 2s 4ms/step - loss: 3.0001e-10 - accuracy: 1.0000 - val_loss: 0.2001 - val_accuracy: 0.9862 Epoch 258/500 469/469 [==============================] - 2s 4ms/step - loss: 2.9405e-10 - accuracy: 1.0000 - val_loss: 0.2001 - val_accuracy: 0.9862 Epoch 259/500 469/469 [==============================] - 2s 4ms/step - loss: 2.9206e-10 - accuracy: 1.0000 - val_loss: 0.2002 - val_accuracy: 0.9862 Epoch 260/500 469/469 [==============================] - 2s 4ms/step - loss: 2.9008e-10 - accuracy: 1.0000 - val_loss: 0.2002 - val_accuracy: 0.9862 Epoch 261/500 469/469 [==============================] - 2s 4ms/step - loss: 2.8809e-10 - accuracy: 1.0000 - val_loss: 0.2003 - val_accuracy: 0.9862 Epoch 262/500 469/469 [==============================] - 2s 4ms/step - loss: 2.8610e-10 - accuracy: 1.0000 - val_loss: 0.2004 - val_accuracy: 0.9862 Epoch 263/500 469/469 [==============================] - 2s 4ms/step - loss: 2.8412e-10 - accuracy: 1.0000 - val_loss: 0.2004 - val_accuracy: 0.9862 Epoch 264/500 469/469 [==============================] - 2s 4ms/step - loss: 2.8213e-10 - accuracy: 1.0000 - val_loss: 0.2005 - val_accuracy: 0.9862 Epoch 265/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7815e-10 - accuracy: 1.0000 - val_loss: 0.2005 - val_accuracy: 0.9862 Epoch 266/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7815e-10 - accuracy: 1.0000 - val_loss: 0.2006 - val_accuracy: 0.9862 Epoch 267/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7815e-10 - accuracy: 1.0000 - val_loss: 0.2006 - val_accuracy: 0.9862 Epoch 268/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7815e-10 - accuracy: 1.0000 - val_loss: 0.2007 - val_accuracy: 0.9862 Epoch 269/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7418e-10 - accuracy: 1.0000 - val_loss: 0.2007 - val_accuracy: 0.9862 Epoch 270/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7021e-10 - accuracy: 1.0000 - val_loss: 0.2008 - val_accuracy: 0.9862 Epoch 271/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7021e-10 - accuracy: 1.0000 - val_loss: 0.2008 - val_accuracy: 0.9862 Epoch 272/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7021e-10 - accuracy: 1.0000 - val_loss: 0.2009 - val_accuracy: 0.9862 Epoch 273/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7021e-10 - accuracy: 1.0000 - val_loss: 0.2009 - val_accuracy: 0.9862 Epoch 274/500 469/469 [==============================] - 2s 4ms/step - loss: 2.6822e-10 - accuracy: 1.0000 - val_loss: 0.2010 - val_accuracy: 0.9862 Epoch 275/500 469/469 [==============================] - 2s 4ms/step - loss: 2.6425e-10 - accuracy: 1.0000 - val_loss: 0.2010 - val_accuracy: 0.9862 Epoch 276/500 469/469 [==============================] - 2s 4ms/step - loss: 2.6425e-10 - accuracy: 1.0000 - val_loss: 0.2011 - val_accuracy: 0.9862 Epoch 277/500 469/469 [==============================] - 2s 4ms/step - loss: 2.6226e-10 - accuracy: 1.0000 - val_loss: 0.2011 - val_accuracy: 0.9862 Epoch 278/500 469/469 [==============================] - 2s 4ms/step - loss: 2.5431e-10 - accuracy: 1.0000 - val_loss: 0.2012 - val_accuracy: 0.9862 Epoch 279/500 469/469 [==============================] - 2s 4ms/step - loss: 2.5233e-10 - accuracy: 1.0000 - val_loss: 0.2012 - val_accuracy: 0.9862 Epoch 280/500 469/469 [==============================] - 2s 4ms/step - loss: 2.5034e-10 - accuracy: 1.0000 - val_loss: 0.2013 - val_accuracy: 0.9862 Epoch 281/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4835e-10 - accuracy: 1.0000 - val_loss: 0.2013 - val_accuracy: 0.9862 Epoch 282/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4438e-10 - accuracy: 1.0000 - val_loss: 0.2014 - val_accuracy: 0.9862 Epoch 283/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4438e-10 - accuracy: 1.0000 - val_loss: 0.2014 - val_accuracy: 0.9862 Epoch 284/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4239e-10 - accuracy: 1.0000 - val_loss: 0.2015 - val_accuracy: 0.9862 Epoch 285/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4239e-10 - accuracy: 1.0000 - val_loss: 0.2015 - val_accuracy: 0.9862 Epoch 286/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4041e-10 - accuracy: 1.0000 - val_loss: 0.2016 - val_accuracy: 0.9862 Epoch 287/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4239e-10 - accuracy: 1.0000 - val_loss: 0.2016 - val_accuracy: 0.9862 Epoch 288/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3842e-10 - accuracy: 1.0000 - val_loss: 0.2017 - val_accuracy: 0.9862 Epoch 289/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3643e-10 - accuracy: 1.0000 - val_loss: 0.2017 - val_accuracy: 0.9862 Epoch 290/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3643e-10 - accuracy: 1.0000 - val_loss: 0.2017 - val_accuracy: 0.9862 Epoch 291/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3444e-10 - accuracy: 1.0000 - val_loss: 0.2018 - val_accuracy: 0.9862 Epoch 292/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3047e-10 - accuracy: 1.0000 - val_loss: 0.2018 - val_accuracy: 0.9862 Epoch 293/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3047e-10 - accuracy: 1.0000 - val_loss: 0.2019 - val_accuracy: 0.9862 Epoch 294/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3047e-10 - accuracy: 1.0000 - val_loss: 0.2019 - val_accuracy: 0.9862 Epoch 295/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3047e-10 - accuracy: 1.0000 - val_loss: 0.2020 - val_accuracy: 0.9862 Epoch 296/500 469/469 [==============================] - 2s 4ms/step - loss: 2.3047e-10 - accuracy: 1.0000 - val_loss: 0.2020 - val_accuracy: 0.9862 Epoch 297/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2848e-10 - accuracy: 1.0000 - val_loss: 0.2021 - val_accuracy: 0.9862 Epoch 298/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2650e-10 - accuracy: 1.0000 - val_loss: 0.2021 - val_accuracy: 0.9862 Epoch 299/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2451e-10 - accuracy: 1.0000 - val_loss: 0.2021 - val_accuracy: 0.9862 Epoch 300/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2054e-10 - accuracy: 1.0000 - val_loss: 0.2022 - val_accuracy: 0.9862 Epoch 301/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2054e-10 - accuracy: 1.0000 - val_loss: 0.2022 - val_accuracy: 0.9862 Epoch 302/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2054e-10 - accuracy: 1.0000 - val_loss: 0.2023 - val_accuracy: 0.9862 Epoch 303/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2054e-10 - accuracy: 1.0000 - val_loss: 0.2023 - val_accuracy: 0.9862 Epoch 304/500 469/469 [==============================] - 2s 4ms/step - loss: 2.2054e-10 - accuracy: 1.0000 - val_loss: 0.2024 - val_accuracy: 0.9862 Epoch 305/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1656e-10 - accuracy: 1.0000 - val_loss: 0.2024 - val_accuracy: 0.9862 Epoch 306/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1458e-10 - accuracy: 1.0000 - val_loss: 0.2024 - val_accuracy: 0.9862 Epoch 307/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1458e-10 - accuracy: 1.0000 - val_loss: 0.2025 - val_accuracy: 0.9862 Epoch 308/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1259e-10 - accuracy: 1.0000 - val_loss: 0.2025 - val_accuracy: 0.9862 Epoch 309/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1259e-10 - accuracy: 1.0000 - val_loss: 0.2026 - val_accuracy: 0.9862 Epoch 310/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1060e-10 - accuracy: 1.0000 - val_loss: 0.2026 - val_accuracy: 0.9862 Epoch 311/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1060e-10 - accuracy: 1.0000 - val_loss: 0.2026 - val_accuracy: 0.9862 Epoch 312/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0862e-10 - accuracy: 1.0000 - val_loss: 0.2027 - val_accuracy: 0.9862 Epoch 313/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0663e-10 - accuracy: 1.0000 - val_loss: 0.2027 - val_accuracy: 0.9862 Epoch 314/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0663e-10 - accuracy: 1.0000 - val_loss: 0.2028 - val_accuracy: 0.9862 Epoch 315/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0464e-10 - accuracy: 1.0000 - val_loss: 0.2028 - val_accuracy: 0.9862 Epoch 316/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0464e-10 - accuracy: 1.0000 - val_loss: 0.2028 - val_accuracy: 0.9862 Epoch 317/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0464e-10 - accuracy: 1.0000 - val_loss: 0.2029 - val_accuracy: 0.9862 Epoch 318/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0464e-10 - accuracy: 1.0000 - val_loss: 0.2029 - val_accuracy: 0.9862 Epoch 319/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0464e-10 - accuracy: 1.0000 - val_loss: 0.2029 - val_accuracy: 0.9862 Epoch 320/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0067e-10 - accuracy: 1.0000 - val_loss: 0.2030 - val_accuracy: 0.9862 Epoch 321/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0067e-10 - accuracy: 1.0000 - val_loss: 0.2030 - val_accuracy: 0.9862 Epoch 322/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-10 - accuracy: 1.0000 - val_loss: 0.2031 - val_accuracy: 0.9862 Epoch 323/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-10 - accuracy: 1.0000 - val_loss: 0.2031 - val_accuracy: 0.9862 Epoch 324/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9272e-10 - accuracy: 1.0000 - val_loss: 0.2031 - val_accuracy: 0.9862 Epoch 325/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9073e-10 - accuracy: 1.0000 - val_loss: 0.2032 - val_accuracy: 0.9862 Epoch 326/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8875e-10 - accuracy: 1.0000 - val_loss: 0.2032 - val_accuracy: 0.9862 Epoch 327/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8875e-10 - accuracy: 1.0000 - val_loss: 0.2032 - val_accuracy: 0.9862 Epoch 328/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8676e-10 - accuracy: 1.0000 - val_loss: 0.2033 - val_accuracy: 0.9862 Epoch 329/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8676e-10 - accuracy: 1.0000 - val_loss: 0.2033 - val_accuracy: 0.9862 Epoch 330/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8477e-10 - accuracy: 1.0000 - val_loss: 0.2033 - val_accuracy: 0.9862 Epoch 331/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8477e-10 - accuracy: 1.0000 - val_loss: 0.2034 - val_accuracy: 0.9862 Epoch 332/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8477e-10 - accuracy: 1.0000 - val_loss: 0.2034 - val_accuracy: 0.9862 Epoch 333/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8477e-10 - accuracy: 1.0000 - val_loss: 0.2034 - val_accuracy: 0.9862 Epoch 334/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8279e-10 - accuracy: 1.0000 - val_loss: 0.2035 - val_accuracy: 0.9862 Epoch 335/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8279e-10 - accuracy: 1.0000 - val_loss: 0.2035 - val_accuracy: 0.9862 Epoch 336/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7881e-10 - accuracy: 1.0000 - val_loss: 0.2036 - val_accuracy: 0.9862 Epoch 337/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7881e-10 - accuracy: 1.0000 - val_loss: 0.2036 - val_accuracy: 0.9862 Epoch 338/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7881e-10 - accuracy: 1.0000 - val_loss: 0.2036 - val_accuracy: 0.9862 Epoch 339/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7683e-10 - accuracy: 1.0000 - val_loss: 0.2037 - val_accuracy: 0.9862 Epoch 340/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7484e-10 - accuracy: 1.0000 - val_loss: 0.2037 - val_accuracy: 0.9862 Epoch 341/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7484e-10 - accuracy: 1.0000 - val_loss: 0.2037 - val_accuracy: 0.9862 Epoch 342/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7484e-10 - accuracy: 1.0000 - val_loss: 0.2038 - val_accuracy: 0.9862 Epoch 343/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7285e-10 - accuracy: 1.0000 - val_loss: 0.2038 - val_accuracy: 0.9862 Epoch 344/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7285e-10 - accuracy: 1.0000 - val_loss: 0.2038 - val_accuracy: 0.9862 Epoch 345/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7087e-10 - accuracy: 1.0000 - val_loss: 0.2039 - val_accuracy: 0.9862 Epoch 346/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6888e-10 - accuracy: 1.0000 - val_loss: 0.2039 - val_accuracy: 0.9862 Epoch 347/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6292e-10 - accuracy: 1.0000 - val_loss: 0.2039 - val_accuracy: 0.9862 Epoch 348/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2039 - val_accuracy: 0.9862 Epoch 349/500 469/469 [==============================] - 2s 5ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2040 - val_accuracy: 0.9862 Epoch 350/500 469/469 [==============================] - 2s 5ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2040 - val_accuracy: 0.9862 Epoch 351/500 469/469 [==============================] - 2s 5ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2040 - val_accuracy: 0.9862 Epoch 352/500 469/469 [==============================] - 2s 5ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2041 - val_accuracy: 0.9862 Epoch 353/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2041 - val_accuracy: 0.9862 Epoch 354/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2041 - val_accuracy: 0.9862 Epoch 355/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2042 - val_accuracy: 0.9862 Epoch 356/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2042 - val_accuracy: 0.9862 Epoch 357/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2042 - val_accuracy: 0.9862 Epoch 358/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2043 - val_accuracy: 0.9862 Epoch 359/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6093e-10 - accuracy: 1.0000 - val_loss: 0.2043 - val_accuracy: 0.9862 Epoch 360/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-10 - accuracy: 1.0000 - val_loss: 0.2043 - val_accuracy: 0.9862 Epoch 361/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-10 - accuracy: 1.0000 - val_loss: 0.2044 - val_accuracy: 0.9862 Epoch 362/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-10 - accuracy: 1.0000 - val_loss: 0.2044 - val_accuracy: 0.9862 Epoch 363/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-10 - accuracy: 1.0000 - val_loss: 0.2044 - val_accuracy: 0.9862 Epoch 364/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-10 - accuracy: 1.0000 - val_loss: 0.2044 - val_accuracy: 0.9862 Epoch 365/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-10 - accuracy: 1.0000 - val_loss: 0.2045 - val_accuracy: 0.9862 Epoch 366/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5696e-10 - accuracy: 1.0000 - val_loss: 0.2045 - val_accuracy: 0.9862 Epoch 367/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5497e-10 - accuracy: 1.0000 - val_loss: 0.2045 - val_accuracy: 0.9862 Epoch 368/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5497e-10 - accuracy: 1.0000 - val_loss: 0.2046 - val_accuracy: 0.9862 Epoch 369/500 469/469 [==============================] - 2s 4ms/step - loss: 1.5100e-10 - accuracy: 1.0000 - val_loss: 0.2046 - val_accuracy: 0.9862 Epoch 370/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4702e-10 - accuracy: 1.0000 - val_loss: 0.2046 - val_accuracy: 0.9862 Epoch 371/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4702e-10 - accuracy: 1.0000 - val_loss: 0.2047 - val_accuracy: 0.9862 Epoch 372/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4702e-10 - accuracy: 1.0000 - val_loss: 0.2047 - val_accuracy: 0.9862 Epoch 373/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4702e-10 - accuracy: 1.0000 - val_loss: 0.2047 - val_accuracy: 0.9862 Epoch 374/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4702e-10 - accuracy: 1.0000 - val_loss: 0.2047 - val_accuracy: 0.9862 Epoch 375/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4702e-10 - accuracy: 1.0000 - val_loss: 0.2048 - val_accuracy: 0.9862 Epoch 376/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4504e-10 - accuracy: 1.0000 - val_loss: 0.2048 - val_accuracy: 0.9862 Epoch 377/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4504e-10 - accuracy: 1.0000 - val_loss: 0.2048 - val_accuracy: 0.9862 Epoch 378/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4504e-10 - accuracy: 1.0000 - val_loss: 0.2048 - val_accuracy: 0.9862 Epoch 379/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4305e-10 - accuracy: 1.0000 - val_loss: 0.2049 - val_accuracy: 0.9862 Epoch 380/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4305e-10 - accuracy: 1.0000 - val_loss: 0.2049 - val_accuracy: 0.9862 Epoch 381/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4305e-10 - accuracy: 1.0000 - val_loss: 0.2049 - val_accuracy: 0.9862 Epoch 382/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4305e-10 - accuracy: 1.0000 - val_loss: 0.2050 - val_accuracy: 0.9862 Epoch 383/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4305e-10 - accuracy: 1.0000 - val_loss: 0.2050 - val_accuracy: 0.9862 Epoch 384/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4305e-10 - accuracy: 1.0000 - val_loss: 0.2050 - val_accuracy: 0.9862 Epoch 385/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4106e-10 - accuracy: 1.0000 - val_loss: 0.2050 - val_accuracy: 0.9862 Epoch 386/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3908e-10 - accuracy: 1.0000 - val_loss: 0.2051 - val_accuracy: 0.9862 Epoch 387/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3709e-10 - accuracy: 1.0000 - val_loss: 0.2051 - val_accuracy: 0.9862 Epoch 388/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3709e-10 - accuracy: 1.0000 - val_loss: 0.2051 - val_accuracy: 0.9862 Epoch 389/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3709e-10 - accuracy: 1.0000 - val_loss: 0.2051 - val_accuracy: 0.9862 Epoch 390/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3510e-10 - accuracy: 1.0000 - val_loss: 0.2052 - val_accuracy: 0.9862 Epoch 391/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3510e-10 - accuracy: 1.0000 - val_loss: 0.2052 - val_accuracy: 0.9862 Epoch 392/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3510e-10 - accuracy: 1.0000 - val_loss: 0.2052 - val_accuracy: 0.9862 Epoch 393/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3312e-10 - accuracy: 1.0000 - val_loss: 0.2052 - val_accuracy: 0.9862 Epoch 394/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3312e-10 - accuracy: 1.0000 - val_loss: 0.2053 - val_accuracy: 0.9862 Epoch 395/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3312e-10 - accuracy: 1.0000 - val_loss: 0.2053 - val_accuracy: 0.9862 Epoch 396/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3312e-10 - accuracy: 1.0000 - val_loss: 0.2053 - val_accuracy: 0.9862 Epoch 397/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3113e-10 - accuracy: 1.0000 - val_loss: 0.2054 - val_accuracy: 0.9862 Epoch 398/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3113e-10 - accuracy: 1.0000 - val_loss: 0.2054 - val_accuracy: 0.9862 Epoch 399/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2054 - val_accuracy: 0.9862 Epoch 400/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2054 - val_accuracy: 0.9862 Epoch 401/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2055 - val_accuracy: 0.9862 Epoch 402/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2055 - val_accuracy: 0.9862 Epoch 403/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2055 - val_accuracy: 0.9862 Epoch 404/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2055 - val_accuracy: 0.9862 Epoch 405/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2056 - val_accuracy: 0.9862 Epoch 406/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2056 - val_accuracy: 0.9862 Epoch 407/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2716e-10 - accuracy: 1.0000 - val_loss: 0.2056 - val_accuracy: 0.9862 Epoch 408/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2056 - val_accuracy: 0.9862 Epoch 409/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2517e-10 - accuracy: 1.0000 - val_loss: 0.2057 - val_accuracy: 0.9862 Epoch 410/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2517e-10 - accuracy: 1.0000 - val_loss: 0.2057 - val_accuracy: 0.9862 Epoch 411/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2318e-10 - accuracy: 1.0000 - val_loss: 0.2057 - val_accuracy: 0.9862 Epoch 412/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2318e-10 - accuracy: 1.0000 - val_loss: 0.2057 - val_accuracy: 0.9862 Epoch 413/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2318e-10 - accuracy: 1.0000 - val_loss: 0.2058 - val_accuracy: 0.9862 Epoch 414/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2058 - val_accuracy: 0.9862 Epoch 415/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2058 - val_accuracy: 0.9862 Epoch 416/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2058 - val_accuracy: 0.9862 Epoch 417/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2058 - val_accuracy: 0.9862 Epoch 418/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2059 - val_accuracy: 0.9862 Epoch 419/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2059 - val_accuracy: 0.9862 Epoch 420/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2059 - val_accuracy: 0.9862 Epoch 421/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2059 - val_accuracy: 0.9862 Epoch 422/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2060 - val_accuracy: 0.9862 Epoch 423/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2060 - val_accuracy: 0.9862 Epoch 424/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2060 - val_accuracy: 0.9862 Epoch 425/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2060 - val_accuracy: 0.9862 Epoch 426/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2061 - val_accuracy: 0.9862 Epoch 427/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2061 - val_accuracy: 0.9862 Epoch 428/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2061 - val_accuracy: 0.9862 Epoch 429/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2061 - val_accuracy: 0.9862 Epoch 430/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2062 - val_accuracy: 0.9862 Epoch 431/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2120e-10 - accuracy: 1.0000 - val_loss: 0.2062 - val_accuracy: 0.9862 Epoch 432/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1921e-10 - accuracy: 1.0000 - val_loss: 0.2062 - val_accuracy: 0.9862 Epoch 433/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1722e-10 - accuracy: 1.0000 - val_loss: 0.2062 - val_accuracy: 0.9862 Epoch 434/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1524e-10 - accuracy: 1.0000 - val_loss: 0.2062 - val_accuracy: 0.9862 Epoch 435/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1325e-10 - accuracy: 1.0000 - val_loss: 0.2063 - val_accuracy: 0.9862 Epoch 436/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1325e-10 - accuracy: 1.0000 - val_loss: 0.2063 - val_accuracy: 0.9862 Epoch 437/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1325e-10 - accuracy: 1.0000 - val_loss: 0.2063 - val_accuracy: 0.9862 Epoch 438/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2063 - val_accuracy: 0.9862 Epoch 439/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2064 - val_accuracy: 0.9862 Epoch 440/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2064 - val_accuracy: 0.9862 Epoch 441/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2064 - val_accuracy: 0.9862 Epoch 442/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2064 - val_accuracy: 0.9862 Epoch 443/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2064 - val_accuracy: 0.9862 Epoch 444/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2065 - val_accuracy: 0.9862 Epoch 445/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2065 - val_accuracy: 0.9862 Epoch 446/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2065 - val_accuracy: 0.9862 Epoch 447/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2065 - val_accuracy: 0.9862 Epoch 448/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2065 - val_accuracy: 0.9862 Epoch 449/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2066 - val_accuracy: 0.9862 Epoch 450/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2066 - val_accuracy: 0.9862 Epoch 451/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2066 - val_accuracy: 0.9862 Epoch 452/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2066 - val_accuracy: 0.9862 Epoch 453/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2067 - val_accuracy: 0.9862 Epoch 454/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2067 - val_accuracy: 0.9862 Epoch 455/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2067 - val_accuracy: 0.9862 Epoch 456/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0928e-10 - accuracy: 1.0000 - val_loss: 0.2067 - val_accuracy: 0.9862 Epoch 457/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0928e-10 - accuracy: 1.0000 - val_loss: 0.2067 - val_accuracy: 0.9862 Epoch 458/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0928e-10 - accuracy: 1.0000 - val_loss: 0.2068 - val_accuracy: 0.9861 Epoch 459/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0729e-10 - accuracy: 1.0000 - val_loss: 0.2068 - val_accuracy: 0.9861 Epoch 460/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0530e-10 - accuracy: 1.0000 - val_loss: 0.2068 - val_accuracy: 0.9861 Epoch 461/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0530e-10 - accuracy: 1.0000 - val_loss: 0.2068 - val_accuracy: 0.9861 Epoch 462/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0530e-10 - accuracy: 1.0000 - val_loss: 0.2068 - val_accuracy: 0.9861 Epoch 463/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0530e-10 - accuracy: 1.0000 - val_loss: 0.2069 - val_accuracy: 0.9861 Epoch 464/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0530e-10 - accuracy: 1.0000 - val_loss: 0.2069 - val_accuracy: 0.9861 Epoch 465/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0530e-10 - accuracy: 1.0000 - val_loss: 0.2069 - val_accuracy: 0.9861 Epoch 466/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0530e-10 - accuracy: 1.0000 - val_loss: 0.2069 - val_accuracy: 0.9861 Epoch 467/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2069 - val_accuracy: 0.9861 Epoch 468/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2070 - val_accuracy: 0.9861 Epoch 469/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2070 - val_accuracy: 0.9861 Epoch 470/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2070 - val_accuracy: 0.9861 Epoch 471/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2070 - val_accuracy: 0.9861 Epoch 472/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2070 - val_accuracy: 0.9861 Epoch 473/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9861 Epoch 474/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0331e-10 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9861 Epoch 475/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0133e-10 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9861 Epoch 476/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0133e-10 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9861 Epoch 477/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0133e-10 - accuracy: 1.0000 - val_loss: 0.2071 - val_accuracy: 0.9861 Epoch 478/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0133e-10 - accuracy: 1.0000 - val_loss: 0.2072 - val_accuracy: 0.9861 Epoch 479/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0133e-10 - accuracy: 1.0000 - val_loss: 0.2072 - val_accuracy: 0.9861 Epoch 480/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0133e-10 - accuracy: 1.0000 - val_loss: 0.2072 - val_accuracy: 0.9861 Epoch 481/500 469/469 [==============================] - 2s 4ms/step - loss: 1.0133e-10 - accuracy: 1.0000 - val_loss: 0.2072 - val_accuracy: 0.9861 Epoch 482/500 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-11 - accuracy: 1.0000 - val_loss: 0.2072 - val_accuracy: 0.9861 Epoch 483/500 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-11 - accuracy: 1.0000 - val_loss: 0.2073 - val_accuracy: 0.9861 Epoch 484/500 469/469 [==============================] - 2s 4ms/step - loss: 9.9341e-11 - accuracy: 1.0000 - val_loss: 0.2073 - val_accuracy: 0.9861 Epoch 485/500 469/469 [==============================] - 2s 4ms/step - loss: 9.7354e-11 - accuracy: 1.0000 - val_loss: 0.2073 - val_accuracy: 0.9861 Epoch 486/500 469/469 [==============================] - 2s 4ms/step - loss: 9.7354e-11 - accuracy: 1.0000 - val_loss: 0.2073 - val_accuracy: 0.9861 Epoch 487/500 469/469 [==============================] - 2s 4ms/step - loss: 9.7354e-11 - accuracy: 1.0000 - val_loss: 0.2073 - val_accuracy: 0.9861 Epoch 488/500 469/469 [==============================] - 2s 4ms/step - loss: 9.7354e-11 - accuracy: 1.0000 - val_loss: 0.2073 - val_accuracy: 0.9861 Epoch 489/500 469/469 [==============================] - 2s 4ms/step - loss: 9.5367e-11 - accuracy: 1.0000 - val_loss: 0.2074 - val_accuracy: 0.9861 Epoch 490/500 469/469 [==============================] - 2s 4ms/step - loss: 9.5367e-11 - accuracy: 1.0000 - val_loss: 0.2074 - val_accuracy: 0.9861 Epoch 491/500 469/469 [==============================] - 2s 4ms/step - loss: 9.5367e-11 - accuracy: 1.0000 - val_loss: 0.2074 - val_accuracy: 0.9861 Epoch 492/500 469/469 [==============================] - 2s 4ms/step - loss: 9.5367e-11 - accuracy: 1.0000 - val_loss: 0.2074 - val_accuracy: 0.9861 Epoch 493/500 469/469 [==============================] - 2s 4ms/step - loss: 9.5367e-11 - accuracy: 1.0000 - val_loss: 0.2074 - val_accuracy: 0.9861 Epoch 494/500 469/469 [==============================] - 2s 4ms/step - loss: 9.5367e-11 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9861 Epoch 495/500 469/469 [==============================] - 2s 4ms/step - loss: 9.3381e-11 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9861 Epoch 496/500 469/469 [==============================] - 2s 4ms/step - loss: 9.3381e-11 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9861 Epoch 497/500 469/469 [==============================] - 2s 4ms/step - loss: 9.3381e-11 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9861 Epoch 498/500 469/469 [==============================] - 2s 4ms/step - loss: 9.1394e-11 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9861 Epoch 499/500 469/469 [==============================] - 2s 4ms/step - loss: 9.1394e-11 - accuracy: 1.0000 - val_loss: 0.2075 - val_accuracy: 0.9861 Epoch 500/500 469/469 [==============================] - 2s 4ms/step - loss: 9.1394e-11 - accuracy: 1.0000 - val_loss: 0.2076 - val_accuracy: 0.9861
Activation function: the logistic sigmoid function; initialization: Xavier initializer; with dropout rate: 0.2 for the first layer and 0.5 for the other hidden layers¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.GlorotNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dropout(0.2),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='sigmoid',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_2" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_2 (Flatten) (None, 784) 0 _________________________________________________________________ dropout_6 (Dropout) (None, 784) 0 _________________________________________________________________ dense_12 (Dense) (None, 1024) 803840 _________________________________________________________________ dropout_7 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_13 (Dense) (None, 1024) 1049600 _________________________________________________________________ dropout_8 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_14 (Dense) (None, 1024) 1049600 _________________________________________________________________ dropout_9 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_15 (Dense) (None, 1024) 1049600 _________________________________________________________________ dropout_10 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_16 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_17 (Dense) (None, 10) 10250 ================================================================= Total params: 5,012,490 Trainable params: 5,012,490 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam()
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"])
historyd = model.fit(X_train, y_train, batch_size=128, epochs=500, validation_data=(X_test, y_test))
Epoch 1/500 469/469 [==============================] - 2s 5ms/step - loss: 1.1013 - accuracy: 0.6043 - val_loss: 0.3344 - val_accuracy: 0.8984 Epoch 2/500 469/469 [==============================] - 2s 4ms/step - loss: 0.3512 - accuracy: 0.8927 - val_loss: 0.2051 - val_accuracy: 0.9374 Epoch 3/500 469/469 [==============================] - 2s 4ms/step - loss: 0.2577 - accuracy: 0.9219 - val_loss: 0.1625 - val_accuracy: 0.9520 Epoch 4/500 469/469 [==============================] - 2s 4ms/step - loss: 0.2136 - accuracy: 0.9355 - val_loss: 0.1271 - val_accuracy: 0.9629 Epoch 5/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1861 - accuracy: 0.9447 - val_loss: 0.1187 - val_accuracy: 0.9653 Epoch 6/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1641 - accuracy: 0.9511 - val_loss: 0.1132 - val_accuracy: 0.9693 Epoch 7/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1524 - accuracy: 0.9538 - val_loss: 0.1031 - val_accuracy: 0.9699 Epoch 8/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1409 - accuracy: 0.9574 - val_loss: 0.0876 - val_accuracy: 0.9741 Epoch 9/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1251 - accuracy: 0.9627 - val_loss: 0.0936 - val_accuracy: 0.9740 Epoch 10/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1147 - accuracy: 0.9654 - val_loss: 0.0881 - val_accuracy: 0.9740 Epoch 11/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1119 - accuracy: 0.9659 - val_loss: 0.0810 - val_accuracy: 0.9764 Epoch 12/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1046 - accuracy: 0.9675 - val_loss: 0.0753 - val_accuracy: 0.9778 Epoch 13/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1007 - accuracy: 0.9694 - val_loss: 0.0731 - val_accuracy: 0.9783 Epoch 14/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0907 - accuracy: 0.9729 - val_loss: 0.0704 - val_accuracy: 0.9797 Epoch 15/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0894 - accuracy: 0.9726 - val_loss: 0.0702 - val_accuracy: 0.9800 Epoch 16/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0843 - accuracy: 0.9737 - val_loss: 0.0686 - val_accuracy: 0.9801 Epoch 17/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0777 - accuracy: 0.9758 - val_loss: 0.0705 - val_accuracy: 0.9811 Epoch 18/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0786 - accuracy: 0.9760 - val_loss: 0.0665 - val_accuracy: 0.9818 Epoch 19/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0757 - accuracy: 0.9771 - val_loss: 0.0637 - val_accuracy: 0.9820 Epoch 20/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0727 - accuracy: 0.9771 - val_loss: 0.0689 - val_accuracy: 0.9821 Epoch 21/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0698 - accuracy: 0.9791 - val_loss: 0.0697 - val_accuracy: 0.9810 Epoch 22/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0664 - accuracy: 0.9793 - val_loss: 0.0651 - val_accuracy: 0.9832 Epoch 23/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0628 - accuracy: 0.9808 - val_loss: 0.0693 - val_accuracy: 0.9826 Epoch 24/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0616 - accuracy: 0.9809 - val_loss: 0.0619 - val_accuracy: 0.9832 Epoch 25/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0578 - accuracy: 0.9825 - val_loss: 0.0642 - val_accuracy: 0.9837 Epoch 26/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0583 - accuracy: 0.9822 - val_loss: 0.0593 - val_accuracy: 0.9838 Epoch 27/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0564 - accuracy: 0.9827 - val_loss: 0.0607 - val_accuracy: 0.9835 Epoch 28/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0549 - accuracy: 0.9834 - val_loss: 0.0602 - val_accuracy: 0.9842 Epoch 29/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0530 - accuracy: 0.9838 - val_loss: 0.0587 - val_accuracy: 0.9850 Epoch 30/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0521 - accuracy: 0.9838 - val_loss: 0.0580 - val_accuracy: 0.9845 Epoch 31/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0507 - accuracy: 0.9842 - val_loss: 0.0573 - val_accuracy: 0.9854 Epoch 32/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0500 - accuracy: 0.9839 - val_loss: 0.0627 - val_accuracy: 0.9844 Epoch 33/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0503 - accuracy: 0.9846 - val_loss: 0.0554 - val_accuracy: 0.9862 Epoch 34/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0464 - accuracy: 0.9855 - val_loss: 0.0617 - val_accuracy: 0.9845 Epoch 35/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0471 - accuracy: 0.9855 - val_loss: 0.0586 - val_accuracy: 0.9856 Epoch 36/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0449 - accuracy: 0.9859 - val_loss: 0.0551 - val_accuracy: 0.9859 Epoch 37/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0452 - accuracy: 0.9861 - val_loss: 0.0558 - val_accuracy: 0.9860 Epoch 38/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0434 - accuracy: 0.9865 - val_loss: 0.0569 - val_accuracy: 0.9859 Epoch 39/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0446 - accuracy: 0.9864 - val_loss: 0.0585 - val_accuracy: 0.9843 Epoch 40/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0416 - accuracy: 0.9868 - val_loss: 0.0621 - val_accuracy: 0.9846 Epoch 41/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0421 - accuracy: 0.9866 - val_loss: 0.0608 - val_accuracy: 0.9852 Epoch 42/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0386 - accuracy: 0.9878 - val_loss: 0.0635 - val_accuracy: 0.9859 Epoch 43/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0383 - accuracy: 0.9880 - val_loss: 0.0626 - val_accuracy: 0.9857 Epoch 44/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0385 - accuracy: 0.9875 - val_loss: 0.0574 - val_accuracy: 0.9858 Epoch 45/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0368 - accuracy: 0.9882 - val_loss: 0.0618 - val_accuracy: 0.9865 Epoch 46/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0381 - accuracy: 0.9882 - val_loss: 0.0541 - val_accuracy: 0.9863 Epoch 47/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0359 - accuracy: 0.9891 - val_loss: 0.0599 - val_accuracy: 0.9856 Epoch 48/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0366 - accuracy: 0.9889 - val_loss: 0.0577 - val_accuracy: 0.9855 Epoch 49/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0336 - accuracy: 0.9899 - val_loss: 0.0603 - val_accuracy: 0.9851 Epoch 50/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0352 - accuracy: 0.9890 - val_loss: 0.0583 - val_accuracy: 0.9865 Epoch 51/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0350 - accuracy: 0.9894 - val_loss: 0.0586 - val_accuracy: 0.9862 Epoch 52/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0335 - accuracy: 0.9898 - val_loss: 0.0645 - val_accuracy: 0.9865 Epoch 53/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0343 - accuracy: 0.9895 - val_loss: 0.0581 - val_accuracy: 0.9867 Epoch 54/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0322 - accuracy: 0.9899 - val_loss: 0.0548 - val_accuracy: 0.9866 Epoch 55/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0337 - accuracy: 0.9890 - val_loss: 0.0555 - val_accuracy: 0.9867 Epoch 56/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0309 - accuracy: 0.9901 - val_loss: 0.0571 - val_accuracy: 0.9865 Epoch 57/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0314 - accuracy: 0.9908 - val_loss: 0.0586 - val_accuracy: 0.9854 Epoch 58/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0302 - accuracy: 0.9908 - val_loss: 0.0568 - val_accuracy: 0.9869 Epoch 59/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0306 - accuracy: 0.9903 - val_loss: 0.0562 - val_accuracy: 0.9872 Epoch 60/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0319 - accuracy: 0.9904 - val_loss: 0.0565 - val_accuracy: 0.9866 Epoch 61/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0291 - accuracy: 0.9906 - val_loss: 0.0598 - val_accuracy: 0.9862 Epoch 62/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0293 - accuracy: 0.9908 - val_loss: 0.0639 - val_accuracy: 0.9873 Epoch 63/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0287 - accuracy: 0.9905 - val_loss: 0.0631 - val_accuracy: 0.9863 Epoch 64/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0294 - accuracy: 0.9908 - val_loss: 0.0594 - val_accuracy: 0.9871 Epoch 65/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0293 - accuracy: 0.9913 - val_loss: 0.0565 - val_accuracy: 0.9866 Epoch 66/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0264 - accuracy: 0.9919 - val_loss: 0.0556 - val_accuracy: 0.9879 Epoch 67/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0289 - accuracy: 0.9910 - val_loss: 0.0555 - val_accuracy: 0.9863 Epoch 68/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0270 - accuracy: 0.9917 - val_loss: 0.0607 - val_accuracy: 0.9873 Epoch 69/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0274 - accuracy: 0.9919 - val_loss: 0.0572 - val_accuracy: 0.9866 Epoch 70/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0266 - accuracy: 0.9914 - val_loss: 0.0597 - val_accuracy: 0.9860 Epoch 71/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0278 - accuracy: 0.9912 - val_loss: 0.0592 - val_accuracy: 0.9872 Epoch 72/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0280 - accuracy: 0.9916 - val_loss: 0.0551 - val_accuracy: 0.9878 Epoch 73/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0250 - accuracy: 0.9922 - val_loss: 0.0610 - val_accuracy: 0.9877 Epoch 74/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0291 - accuracy: 0.9913 - val_loss: 0.0553 - val_accuracy: 0.9872 Epoch 75/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0243 - accuracy: 0.9926 - val_loss: 0.0604 - val_accuracy: 0.9877 Epoch 76/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0255 - accuracy: 0.9923 - val_loss: 0.0571 - val_accuracy: 0.9868 Epoch 77/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0260 - accuracy: 0.9919 - val_loss: 0.0569 - val_accuracy: 0.9869 Epoch 78/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0256 - accuracy: 0.9925 - val_loss: 0.0554 - val_accuracy: 0.9873 Epoch 79/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0248 - accuracy: 0.9927 - val_loss: 0.0550 - val_accuracy: 0.9884 Epoch 80/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0252 - accuracy: 0.9922 - val_loss: 0.0595 - val_accuracy: 0.9874 Epoch 81/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0257 - accuracy: 0.9916 - val_loss: 0.0654 - val_accuracy: 0.9868 Epoch 82/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0273 - accuracy: 0.9911 - val_loss: 0.0580 - val_accuracy: 0.9867 Epoch 83/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0248 - accuracy: 0.9926 - val_loss: 0.0614 - val_accuracy: 0.9870 Epoch 84/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0246 - accuracy: 0.9924 - val_loss: 0.0590 - val_accuracy: 0.9878 Epoch 85/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0241 - accuracy: 0.9927 - val_loss: 0.0525 - val_accuracy: 0.9874 Epoch 86/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0224 - accuracy: 0.9930 - val_loss: 0.0633 - val_accuracy: 0.9876 Epoch 87/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0253 - accuracy: 0.9922 - val_loss: 0.0570 - val_accuracy: 0.9877 Epoch 88/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0243 - accuracy: 0.9925 - val_loss: 0.0562 - val_accuracy: 0.9879 Epoch 89/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0239 - accuracy: 0.9927 - val_loss: 0.0537 - val_accuracy: 0.9887 Epoch 90/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0240 - accuracy: 0.9924 - val_loss: 0.0562 - val_accuracy: 0.9881 Epoch 91/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0218 - accuracy: 0.9931 - val_loss: 0.0528 - val_accuracy: 0.9889 Epoch 92/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0224 - accuracy: 0.9932 - val_loss: 0.0543 - val_accuracy: 0.9880 Epoch 93/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0238 - accuracy: 0.9929 - val_loss: 0.0528 - val_accuracy: 0.9882 Epoch 94/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0231 - accuracy: 0.9933 - val_loss: 0.0582 - val_accuracy: 0.9884 Epoch 95/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0232 - accuracy: 0.9927 - val_loss: 0.0584 - val_accuracy: 0.9880 Epoch 96/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0226 - accuracy: 0.9927 - val_loss: 0.0593 - val_accuracy: 0.9880 Epoch 97/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0216 - accuracy: 0.9936 - val_loss: 0.0587 - val_accuracy: 0.9881 Epoch 98/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0226 - accuracy: 0.9932 - val_loss: 0.0593 - val_accuracy: 0.9872 Epoch 99/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0211 - accuracy: 0.9934 - val_loss: 0.0612 - val_accuracy: 0.9869 Epoch 100/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0219 - accuracy: 0.9931 - val_loss: 0.0565 - val_accuracy: 0.9877 Epoch 101/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0227 - accuracy: 0.9928 - val_loss: 0.0551 - val_accuracy: 0.9882 Epoch 102/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0223 - accuracy: 0.9930 - val_loss: 0.0524 - val_accuracy: 0.9888 Epoch 103/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0235 - accuracy: 0.9933 - val_loss: 0.0535 - val_accuracy: 0.9882 Epoch 104/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0231 - accuracy: 0.9932 - val_loss: 0.0610 - val_accuracy: 0.9873 Epoch 105/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0212 - accuracy: 0.9934 - val_loss: 0.0553 - val_accuracy: 0.9888 Epoch 106/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0225 - accuracy: 0.9931 - val_loss: 0.0506 - val_accuracy: 0.9897 Epoch 107/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0235 - accuracy: 0.9932 - val_loss: 0.0560 - val_accuracy: 0.9882 Epoch 108/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0193 - accuracy: 0.9941 - val_loss: 0.0602 - val_accuracy: 0.9868 Epoch 109/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0202 - accuracy: 0.9938 - val_loss: 0.0621 - val_accuracy: 0.9874 Epoch 110/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0214 - accuracy: 0.9936 - val_loss: 0.0572 - val_accuracy: 0.9883 Epoch 111/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0194 - accuracy: 0.9937 - val_loss: 0.0628 - val_accuracy: 0.9880 Epoch 112/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0205 - accuracy: 0.9937 - val_loss: 0.0544 - val_accuracy: 0.9890 Epoch 113/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0192 - accuracy: 0.9941 - val_loss: 0.0560 - val_accuracy: 0.9885 Epoch 114/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0198 - accuracy: 0.9943 - val_loss: 0.0554 - val_accuracy: 0.9884 Epoch 115/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0222 - accuracy: 0.9932 - val_loss: 0.0575 - val_accuracy: 0.9886 Epoch 116/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0191 - accuracy: 0.9938 - val_loss: 0.0570 - val_accuracy: 0.9881 Epoch 117/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0192 - accuracy: 0.9941 - val_loss: 0.0621 - val_accuracy: 0.9872 Epoch 118/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0209 - accuracy: 0.9939 - val_loss: 0.0562 - val_accuracy: 0.9882 Epoch 119/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0191 - accuracy: 0.9943 - val_loss: 0.0597 - val_accuracy: 0.9875 Epoch 120/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0204 - accuracy: 0.9941 - val_loss: 0.0573 - val_accuracy: 0.9883 Epoch 121/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0205 - accuracy: 0.9939 - val_loss: 0.0606 - val_accuracy: 0.9878 Epoch 122/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0200 - accuracy: 0.9937 - val_loss: 0.0596 - val_accuracy: 0.9889 Epoch 123/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0190 - accuracy: 0.9945 - val_loss: 0.0598 - val_accuracy: 0.9875 Epoch 124/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0204 - accuracy: 0.9941 - val_loss: 0.0544 - val_accuracy: 0.9892 Epoch 125/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0195 - accuracy: 0.9943 - val_loss: 0.0543 - val_accuracy: 0.9892 Epoch 126/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0182 - accuracy: 0.9944 - val_loss: 0.0563 - val_accuracy: 0.9887 Epoch 127/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0193 - accuracy: 0.9940 - val_loss: 0.0565 - val_accuracy: 0.9881 Epoch 128/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0184 - accuracy: 0.9944 - val_loss: 0.0602 - val_accuracy: 0.9868 Epoch 129/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0203 - accuracy: 0.9939 - val_loss: 0.0606 - val_accuracy: 0.9878 Epoch 130/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0189 - accuracy: 0.9941 - val_loss: 0.0550 - val_accuracy: 0.9889 Epoch 131/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0180 - accuracy: 0.9945 - val_loss: 0.0634 - val_accuracy: 0.9881 Epoch 132/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0176 - accuracy: 0.9949 - val_loss: 0.0592 - val_accuracy: 0.9876 Epoch 133/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0212 - accuracy: 0.9937 - val_loss: 0.0561 - val_accuracy: 0.9882 Epoch 134/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0191 - accuracy: 0.9945 - val_loss: 0.0567 - val_accuracy: 0.9883 Epoch 135/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0199 - accuracy: 0.9942 - val_loss: 0.0571 - val_accuracy: 0.9891 Epoch 136/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0178 - accuracy: 0.9943 - val_loss: 0.0548 - val_accuracy: 0.9896 Epoch 137/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0191 - accuracy: 0.9945 - val_loss: 0.0593 - val_accuracy: 0.9891 Epoch 138/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0185 - accuracy: 0.9944 - val_loss: 0.0660 - val_accuracy: 0.9878 Epoch 139/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0199 - accuracy: 0.9940 - val_loss: 0.0613 - val_accuracy: 0.9887 Epoch 140/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0164 - accuracy: 0.9952 - val_loss: 0.0629 - val_accuracy: 0.9882 Epoch 141/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0183 - accuracy: 0.9946 - val_loss: 0.0587 - val_accuracy: 0.9883 Epoch 142/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0181 - accuracy: 0.9949 - val_loss: 0.0631 - val_accuracy: 0.9875 Epoch 143/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0207 - accuracy: 0.9941 - val_loss: 0.0588 - val_accuracy: 0.9888 Epoch 144/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0174 - accuracy: 0.9952 - val_loss: 0.0634 - val_accuracy: 0.9876 Epoch 145/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0176 - accuracy: 0.9947 - val_loss: 0.0629 - val_accuracy: 0.9884 Epoch 146/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0193 - accuracy: 0.9945 - val_loss: 0.0596 - val_accuracy: 0.9882 Epoch 147/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0178 - accuracy: 0.9944 - val_loss: 0.0598 - val_accuracy: 0.9891 Epoch 148/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0177 - accuracy: 0.9945 - val_loss: 0.0626 - val_accuracy: 0.9882 Epoch 149/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0185 - accuracy: 0.9947 - val_loss: 0.0589 - val_accuracy: 0.9877 Epoch 150/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0179 - accuracy: 0.9947 - val_loss: 0.0621 - val_accuracy: 0.9872 Epoch 151/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0170 - accuracy: 0.9949 - val_loss: 0.0675 - val_accuracy: 0.9880 Epoch 152/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0170 - accuracy: 0.9952 - val_loss: 0.0619 - val_accuracy: 0.9884 Epoch 153/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0174 - accuracy: 0.9950 - val_loss: 0.0613 - val_accuracy: 0.9887 Epoch 154/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0179 - accuracy: 0.9949 - val_loss: 0.0667 - val_accuracy: 0.9871 Epoch 155/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0171 - accuracy: 0.9950 - val_loss: 0.0616 - val_accuracy: 0.9884 Epoch 156/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0158 - accuracy: 0.9953 - val_loss: 0.0638 - val_accuracy: 0.9884 Epoch 157/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0191 - accuracy: 0.9944 - val_loss: 0.0623 - val_accuracy: 0.9887 Epoch 158/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0172 - accuracy: 0.9948 - val_loss: 0.0641 - val_accuracy: 0.9888 Epoch 159/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0164 - accuracy: 0.9952 - val_loss: 0.0657 - val_accuracy: 0.9885 Epoch 160/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0161 - accuracy: 0.9949 - val_loss: 0.0636 - val_accuracy: 0.9885 Epoch 161/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0201 - accuracy: 0.9941 - val_loss: 0.0618 - val_accuracy: 0.9878 Epoch 162/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0161 - accuracy: 0.9952 - val_loss: 0.0635 - val_accuracy: 0.9893 Epoch 163/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0169 - accuracy: 0.9951 - val_loss: 0.0635 - val_accuracy: 0.9870 Epoch 164/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0172 - accuracy: 0.9948 - val_loss: 0.0581 - val_accuracy: 0.9889 Epoch 165/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0182 - accuracy: 0.9948 - val_loss: 0.0589 - val_accuracy: 0.9879 Epoch 166/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0163 - accuracy: 0.9954 - val_loss: 0.0598 - val_accuracy: 0.9889 Epoch 167/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0165 - accuracy: 0.9952 - val_loss: 0.0649 - val_accuracy: 0.9878 Epoch 168/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0180 - accuracy: 0.9948 - val_loss: 0.0585 - val_accuracy: 0.9878 Epoch 169/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0171 - accuracy: 0.9949 - val_loss: 0.0589 - val_accuracy: 0.9880 Epoch 170/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0156 - accuracy: 0.9955 - val_loss: 0.0602 - val_accuracy: 0.9879 Epoch 171/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0156 - accuracy: 0.9956 - val_loss: 0.0586 - val_accuracy: 0.9886 Epoch 172/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0163 - accuracy: 0.9951 - val_loss: 0.0577 - val_accuracy: 0.9888 Epoch 173/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0190 - accuracy: 0.9946 - val_loss: 0.0551 - val_accuracy: 0.9896 Epoch 174/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0155 - accuracy: 0.9954 - val_loss: 0.0565 - val_accuracy: 0.9894 Epoch 175/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0160 - accuracy: 0.9952 - val_loss: 0.0600 - val_accuracy: 0.9880 Epoch 176/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0163 - accuracy: 0.9952 - val_loss: 0.0562 - val_accuracy: 0.9892 Epoch 177/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0152 - accuracy: 0.9955 - val_loss: 0.0668 - val_accuracy: 0.9881 Epoch 178/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0171 - accuracy: 0.9948 - val_loss: 0.0642 - val_accuracy: 0.9870 Epoch 179/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0164 - accuracy: 0.9951 - val_loss: 0.0617 - val_accuracy: 0.9893 Epoch 180/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0162 - accuracy: 0.9954 - val_loss: 0.0610 - val_accuracy: 0.9892 Epoch 181/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0140 - accuracy: 0.9959 - val_loss: 0.0624 - val_accuracy: 0.9881 Epoch 182/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0167 - accuracy: 0.9951 - val_loss: 0.0650 - val_accuracy: 0.9874 Epoch 183/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0167 - accuracy: 0.9951 - val_loss: 0.0575 - val_accuracy: 0.9890 Epoch 184/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0157 - accuracy: 0.9954 - val_loss: 0.0606 - val_accuracy: 0.9879 Epoch 185/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0175 - accuracy: 0.9951 - val_loss: 0.0617 - val_accuracy: 0.9886 Epoch 186/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9958 - val_loss: 0.0636 - val_accuracy: 0.9891 Epoch 187/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9962 - val_loss: 0.0676 - val_accuracy: 0.9883 Epoch 188/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0166 - accuracy: 0.9953 - val_loss: 0.0685 - val_accuracy: 0.9875 Epoch 189/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0160 - accuracy: 0.9954 - val_loss: 0.0696 - val_accuracy: 0.9877 Epoch 190/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0160 - accuracy: 0.9954 - val_loss: 0.0652 - val_accuracy: 0.9889 Epoch 191/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0163 - accuracy: 0.9954 - val_loss: 0.0624 - val_accuracy: 0.9891 Epoch 192/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0166 - accuracy: 0.9952 - val_loss: 0.0661 - val_accuracy: 0.9886 Epoch 193/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0173 - accuracy: 0.9949 - val_loss: 0.0680 - val_accuracy: 0.9882 Epoch 194/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0145 - accuracy: 0.9956 - val_loss: 0.0639 - val_accuracy: 0.9889 Epoch 195/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0158 - accuracy: 0.9953 - val_loss: 0.0630 - val_accuracy: 0.9874 Epoch 196/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0175 - accuracy: 0.9949 - val_loss: 0.0599 - val_accuracy: 0.9884 Epoch 197/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0160 - accuracy: 0.9958 - val_loss: 0.0624 - val_accuracy: 0.9883 Epoch 198/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0149 - accuracy: 0.9955 - val_loss: 0.0669 - val_accuracy: 0.9879 Epoch 199/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0147 - accuracy: 0.9957 - val_loss: 0.0632 - val_accuracy: 0.9884 Epoch 200/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0168 - accuracy: 0.9951 - val_loss: 0.0599 - val_accuracy: 0.9880 Epoch 201/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9955 - val_loss: 0.0642 - val_accuracy: 0.9891 Epoch 202/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0168 - accuracy: 0.9954 - val_loss: 0.0592 - val_accuracy: 0.9892 Epoch 203/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0187 - accuracy: 0.9947 - val_loss: 0.0626 - val_accuracy: 0.9879 Epoch 204/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0165 - accuracy: 0.9956 - val_loss: 0.0629 - val_accuracy: 0.9876 Epoch 205/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0148 - accuracy: 0.9956 - val_loss: 0.0643 - val_accuracy: 0.9891 Epoch 206/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0157 - accuracy: 0.9955 - val_loss: 0.0628 - val_accuracy: 0.9880 Epoch 207/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0154 - accuracy: 0.9956 - val_loss: 0.0603 - val_accuracy: 0.9882 Epoch 208/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0177 - accuracy: 0.9949 - val_loss: 0.0616 - val_accuracy: 0.9889 Epoch 209/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0175 - accuracy: 0.9951 - val_loss: 0.0625 - val_accuracy: 0.9892 Epoch 210/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0165 - accuracy: 0.9952 - val_loss: 0.0596 - val_accuracy: 0.9889 Epoch 211/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0156 - accuracy: 0.9954 - val_loss: 0.0618 - val_accuracy: 0.9888 Epoch 212/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0151 - accuracy: 0.9958 - val_loss: 0.0608 - val_accuracy: 0.9889 Epoch 213/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0158 - accuracy: 0.9955 - val_loss: 0.0623 - val_accuracy: 0.9894 Epoch 214/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9961 - val_loss: 0.0604 - val_accuracy: 0.9900 Epoch 215/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0153 - accuracy: 0.9957 - val_loss: 0.0645 - val_accuracy: 0.9888 Epoch 216/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0158 - accuracy: 0.9955 - val_loss: 0.0696 - val_accuracy: 0.9877 Epoch 217/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0154 - accuracy: 0.9959 - val_loss: 0.0645 - val_accuracy: 0.9888 Epoch 218/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0150 - accuracy: 0.9956 - val_loss: 0.0761 - val_accuracy: 0.9881 Epoch 219/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0165 - accuracy: 0.9954 - val_loss: 0.0664 - val_accuracy: 0.9885 Epoch 220/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0164 - accuracy: 0.9951 - val_loss: 0.0669 - val_accuracy: 0.9884 Epoch 221/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0153 - accuracy: 0.9957 - val_loss: 0.0668 - val_accuracy: 0.9886 Epoch 222/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0141 - accuracy: 0.9961 - val_loss: 0.0709 - val_accuracy: 0.9890 Epoch 223/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0147 - accuracy: 0.9959 - val_loss: 0.0690 - val_accuracy: 0.9890 Epoch 224/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0157 - accuracy: 0.9955 - val_loss: 0.0700 - val_accuracy: 0.9883 Epoch 225/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0153 - accuracy: 0.9957 - val_loss: 0.0640 - val_accuracy: 0.9886 Epoch 226/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0153 - accuracy: 0.9953 - val_loss: 0.0650 - val_accuracy: 0.9889 Epoch 227/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9958 - val_loss: 0.0682 - val_accuracy: 0.9888 Epoch 228/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0167 - accuracy: 0.9953 - val_loss: 0.0639 - val_accuracy: 0.9884 Epoch 229/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0160 - accuracy: 0.9956 - val_loss: 0.0621 - val_accuracy: 0.9886 Epoch 230/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0151 - accuracy: 0.9960 - val_loss: 0.0678 - val_accuracy: 0.9882 Epoch 231/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0151 - accuracy: 0.9954 - val_loss: 0.0667 - val_accuracy: 0.9879 Epoch 232/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0145 - accuracy: 0.9959 - val_loss: 0.0630 - val_accuracy: 0.9889 Epoch 233/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0153 - accuracy: 0.9959 - val_loss: 0.0629 - val_accuracy: 0.9884 Epoch 234/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9958 - val_loss: 0.0689 - val_accuracy: 0.9881 Epoch 235/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0159 - accuracy: 0.9954 - val_loss: 0.0628 - val_accuracy: 0.9886 Epoch 236/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0170 - accuracy: 0.9951 - val_loss: 0.0628 - val_accuracy: 0.9885 Epoch 237/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9961 - val_loss: 0.0689 - val_accuracy: 0.9887 Epoch 238/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0145 - accuracy: 0.9959 - val_loss: 0.0626 - val_accuracy: 0.9899 Epoch 239/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9958 - val_loss: 0.0639 - val_accuracy: 0.9897 Epoch 240/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0160 - accuracy: 0.9957 - val_loss: 0.0650 - val_accuracy: 0.9886 Epoch 241/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9958 - val_loss: 0.0602 - val_accuracy: 0.9896 Epoch 242/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9961 - val_loss: 0.0624 - val_accuracy: 0.9895 Epoch 243/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0147 - accuracy: 0.9959 - val_loss: 0.0647 - val_accuracy: 0.9892 Epoch 244/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0150 - accuracy: 0.9955 - val_loss: 0.0605 - val_accuracy: 0.9896 Epoch 245/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9958 - val_loss: 0.0635 - val_accuracy: 0.9893 Epoch 246/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0130 - accuracy: 0.9961 - val_loss: 0.0733 - val_accuracy: 0.9886 Epoch 247/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0158 - accuracy: 0.9956 - val_loss: 0.0661 - val_accuracy: 0.9885 Epoch 248/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9959 - val_loss: 0.0682 - val_accuracy: 0.9892 Epoch 249/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9960 - val_loss: 0.0642 - val_accuracy: 0.9894 Epoch 250/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9965 - val_loss: 0.0650 - val_accuracy: 0.9897 Epoch 251/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0149 - accuracy: 0.9957 - val_loss: 0.0641 - val_accuracy: 0.9892 Epoch 252/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9962 - val_loss: 0.0606 - val_accuracy: 0.9896 Epoch 253/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9964 - val_loss: 0.0662 - val_accuracy: 0.9886 Epoch 254/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9960 - val_loss: 0.0635 - val_accuracy: 0.9889 Epoch 255/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0147 - accuracy: 0.9961 - val_loss: 0.0605 - val_accuracy: 0.9893 Epoch 256/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9956 - val_loss: 0.0646 - val_accuracy: 0.9888 Epoch 257/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0145 - accuracy: 0.9959 - val_loss: 0.0643 - val_accuracy: 0.9887 Epoch 258/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9962 - val_loss: 0.0563 - val_accuracy: 0.9898 Epoch 259/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9958 - val_loss: 0.0664 - val_accuracy: 0.9892 Epoch 260/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9963 - val_loss: 0.0626 - val_accuracy: 0.9894 Epoch 261/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0162 - accuracy: 0.9954 - val_loss: 0.0626 - val_accuracy: 0.9892 Epoch 262/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0146 - accuracy: 0.9958 - val_loss: 0.0654 - val_accuracy: 0.9887 Epoch 263/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0151 - accuracy: 0.9955 - val_loss: 0.0627 - val_accuracy: 0.9883 Epoch 264/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0118 - accuracy: 0.9964 - val_loss: 0.0687 - val_accuracy: 0.9893 Epoch 265/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0174 - accuracy: 0.9953 - val_loss: 0.0610 - val_accuracy: 0.9887 Epoch 266/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9961 - val_loss: 0.0668 - val_accuracy: 0.9882 Epoch 267/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0155 - accuracy: 0.9958 - val_loss: 0.0615 - val_accuracy: 0.9890 Epoch 268/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0147 - accuracy: 0.9957 - val_loss: 0.0664 - val_accuracy: 0.9890 Epoch 269/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0155 - accuracy: 0.9960 - val_loss: 0.0672 - val_accuracy: 0.9883 Epoch 270/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0150 - accuracy: 0.9957 - val_loss: 0.0636 - val_accuracy: 0.9888 Epoch 271/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9965 - val_loss: 0.0684 - val_accuracy: 0.9883 Epoch 272/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9958 - val_loss: 0.0680 - val_accuracy: 0.9882 Epoch 273/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9961 - val_loss: 0.0726 - val_accuracy: 0.9881 Epoch 274/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0166 - accuracy: 0.9955 - val_loss: 0.0640 - val_accuracy: 0.9881 Epoch 275/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9965 - val_loss: 0.0699 - val_accuracy: 0.9884 Epoch 276/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9964 - val_loss: 0.0695 - val_accuracy: 0.9888 Epoch 277/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9961 - val_loss: 0.0729 - val_accuracy: 0.9883 Epoch 278/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0160 - accuracy: 0.9957 - val_loss: 0.0618 - val_accuracy: 0.9884 Epoch 279/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0141 - accuracy: 0.9958 - val_loss: 0.0659 - val_accuracy: 0.9887 Epoch 280/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9958 - val_loss: 0.0704 - val_accuracy: 0.9881 Epoch 281/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9960 - val_loss: 0.0671 - val_accuracy: 0.9899 Epoch 282/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0157 - accuracy: 0.9954 - val_loss: 0.0683 - val_accuracy: 0.9883 Epoch 283/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9959 - val_loss: 0.0675 - val_accuracy: 0.9881 Epoch 284/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0121 - accuracy: 0.9964 - val_loss: 0.0689 - val_accuracy: 0.9891 Epoch 285/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0149 - accuracy: 0.9959 - val_loss: 0.0651 - val_accuracy: 0.9877 Epoch 286/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9959 - val_loss: 0.0615 - val_accuracy: 0.9889 Epoch 287/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0150 - accuracy: 0.9958 - val_loss: 0.0619 - val_accuracy: 0.9886 Epoch 288/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0137 - accuracy: 0.9964 - val_loss: 0.0625 - val_accuracy: 0.9893 Epoch 289/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0143 - accuracy: 0.9957 - val_loss: 0.0705 - val_accuracy: 0.9887 Epoch 290/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0139 - accuracy: 0.9963 - val_loss: 0.0714 - val_accuracy: 0.9889 Epoch 291/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9959 - val_loss: 0.0725 - val_accuracy: 0.9896 Epoch 292/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0157 - accuracy: 0.9956 - val_loss: 0.0653 - val_accuracy: 0.9886 Epoch 293/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9961 - val_loss: 0.0692 - val_accuracy: 0.9886 Epoch 294/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0146 - accuracy: 0.9955 - val_loss: 0.0682 - val_accuracy: 0.9891 Epoch 295/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0137 - accuracy: 0.9961 - val_loss: 0.0625 - val_accuracy: 0.9887 Epoch 296/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9962 - val_loss: 0.0645 - val_accuracy: 0.9895 Epoch 297/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9958 - val_loss: 0.0650 - val_accuracy: 0.9879 Epoch 298/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9962 - val_loss: 0.0690 - val_accuracy: 0.9882 Epoch 299/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9965 - val_loss: 0.0665 - val_accuracy: 0.9885 Epoch 300/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0131 - accuracy: 0.9963 - val_loss: 0.0753 - val_accuracy: 0.9885 Epoch 301/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9961 - val_loss: 0.0634 - val_accuracy: 0.9900 Epoch 302/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9957 - val_loss: 0.0617 - val_accuracy: 0.9895 Epoch 303/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9965 - val_loss: 0.0629 - val_accuracy: 0.9883 Epoch 304/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9960 - val_loss: 0.0601 - val_accuracy: 0.9891 Epoch 305/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9962 - val_loss: 0.0629 - val_accuracy: 0.9880 Epoch 306/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0135 - accuracy: 0.9962 - val_loss: 0.0675 - val_accuracy: 0.9887 Epoch 307/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0118 - accuracy: 0.9965 - val_loss: 0.0755 - val_accuracy: 0.9884 Epoch 308/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9965 - val_loss: 0.0678 - val_accuracy: 0.9889 Epoch 309/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9968 - val_loss: 0.0672 - val_accuracy: 0.9887 Epoch 310/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0145 - accuracy: 0.9961 - val_loss: 0.0647 - val_accuracy: 0.9886 Epoch 311/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0130 - accuracy: 0.9963 - val_loss: 0.0663 - val_accuracy: 0.9881 Epoch 312/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9961 - val_loss: 0.0668 - val_accuracy: 0.9881 Epoch 313/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9960 - val_loss: 0.0646 - val_accuracy: 0.9884 Epoch 314/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9964 - val_loss: 0.0718 - val_accuracy: 0.9880 Epoch 315/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0141 - accuracy: 0.9963 - val_loss: 0.0651 - val_accuracy: 0.9875 Epoch 316/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0151 - accuracy: 0.9956 - val_loss: 0.0713 - val_accuracy: 0.9875 Epoch 317/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0150 - accuracy: 0.9960 - val_loss: 0.0661 - val_accuracy: 0.9882 Epoch 318/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9963 - val_loss: 0.0714 - val_accuracy: 0.9874 Epoch 319/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0146 - accuracy: 0.9963 - val_loss: 0.0658 - val_accuracy: 0.9886 Epoch 320/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9965 - val_loss: 0.0689 - val_accuracy: 0.9890 Epoch 321/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9963 - val_loss: 0.0649 - val_accuracy: 0.9888 Epoch 322/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9960 - val_loss: 0.0646 - val_accuracy: 0.9892 Epoch 323/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9966 - val_loss: 0.0712 - val_accuracy: 0.9885 Epoch 324/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9963 - val_loss: 0.0691 - val_accuracy: 0.9882 Epoch 325/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9963 - val_loss: 0.0736 - val_accuracy: 0.9885 Epoch 326/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0130 - accuracy: 0.9964 - val_loss: 0.0710 - val_accuracy: 0.9896 Epoch 327/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9964 - val_loss: 0.0720 - val_accuracy: 0.9894 Epoch 328/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9965 - val_loss: 0.0627 - val_accuracy: 0.9891 Epoch 329/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9967 - val_loss: 0.0639 - val_accuracy: 0.9890 Epoch 330/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9966 - val_loss: 0.0667 - val_accuracy: 0.9886 Epoch 331/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9965 - val_loss: 0.0615 - val_accuracy: 0.9893 Epoch 332/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0137 - accuracy: 0.9962 - val_loss: 0.0635 - val_accuracy: 0.9890 Epoch 333/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0151 - accuracy: 0.9957 - val_loss: 0.0690 - val_accuracy: 0.9884 Epoch 334/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9960 - val_loss: 0.0629 - val_accuracy: 0.9888 Epoch 335/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0137 - accuracy: 0.9961 - val_loss: 0.0664 - val_accuracy: 0.9885 Epoch 336/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9965 - val_loss: 0.0704 - val_accuracy: 0.9885 Epoch 337/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9966 - val_loss: 0.0620 - val_accuracy: 0.9894 Epoch 338/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0145 - accuracy: 0.9960 - val_loss: 0.0663 - val_accuracy: 0.9894 Epoch 339/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9965 - val_loss: 0.0693 - val_accuracy: 0.9893 Epoch 340/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9966 - val_loss: 0.0643 - val_accuracy: 0.9881 Epoch 341/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9959 - val_loss: 0.0678 - val_accuracy: 0.9885 Epoch 342/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9965 - val_loss: 0.0664 - val_accuracy: 0.9896 Epoch 343/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9966 - val_loss: 0.0699 - val_accuracy: 0.9885 Epoch 344/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9963 - val_loss: 0.0695 - val_accuracy: 0.9882 Epoch 345/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9964 - val_loss: 0.0703 - val_accuracy: 0.9890 Epoch 346/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0155 - accuracy: 0.9962 - val_loss: 0.0695 - val_accuracy: 0.9884 Epoch 347/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0162 - accuracy: 0.9961 - val_loss: 0.0603 - val_accuracy: 0.9897 Epoch 348/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0131 - accuracy: 0.9962 - val_loss: 0.0637 - val_accuracy: 0.9893 Epoch 349/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0131 - accuracy: 0.9961 - val_loss: 0.0654 - val_accuracy: 0.9887 Epoch 350/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0151 - accuracy: 0.9959 - val_loss: 0.0602 - val_accuracy: 0.9891 Epoch 351/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0126 - accuracy: 0.9963 - val_loss: 0.0629 - val_accuracy: 0.9899 Epoch 352/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9960 - val_loss: 0.0658 - val_accuracy: 0.9894 Epoch 353/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9965 - val_loss: 0.0703 - val_accuracy: 0.9890 Epoch 354/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9968 - val_loss: 0.0717 - val_accuracy: 0.9889 Epoch 355/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9965 - val_loss: 0.0687 - val_accuracy: 0.9877 Epoch 356/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9966 - val_loss: 0.0713 - val_accuracy: 0.9893 Epoch 357/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0146 - accuracy: 0.9959 - val_loss: 0.0675 - val_accuracy: 0.9887 Epoch 358/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0146 - accuracy: 0.9960 - val_loss: 0.0610 - val_accuracy: 0.9887 Epoch 359/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9964 - val_loss: 0.0669 - val_accuracy: 0.9891 Epoch 360/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0116 - accuracy: 0.9971 - val_loss: 0.0693 - val_accuracy: 0.9875 Epoch 361/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9966 - val_loss: 0.0675 - val_accuracy: 0.9891 Epoch 362/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0146 - accuracy: 0.9962 - val_loss: 0.0650 - val_accuracy: 0.9890 Epoch 363/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9965 - val_loss: 0.0623 - val_accuracy: 0.9892 Epoch 364/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9963 - val_loss: 0.0674 - val_accuracy: 0.9890 Epoch 365/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9964 - val_loss: 0.0708 - val_accuracy: 0.9893 Epoch 366/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9958 - val_loss: 0.0660 - val_accuracy: 0.9889 Epoch 367/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9967 - val_loss: 0.0705 - val_accuracy: 0.9882 Epoch 368/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0121 - accuracy: 0.9965 - val_loss: 0.0706 - val_accuracy: 0.9889 Epoch 369/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0130 - accuracy: 0.9963 - val_loss: 0.0685 - val_accuracy: 0.9882 Epoch 370/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0097 - accuracy: 0.9970 - val_loss: 0.0752 - val_accuracy: 0.9887 Epoch 371/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9969 - val_loss: 0.0760 - val_accuracy: 0.9882 Epoch 372/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9967 - val_loss: 0.0675 - val_accuracy: 0.9887 Epoch 373/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9966 - val_loss: 0.0661 - val_accuracy: 0.9886 Epoch 374/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9965 - val_loss: 0.0633 - val_accuracy: 0.9892 Epoch 375/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0118 - accuracy: 0.9966 - val_loss: 0.0716 - val_accuracy: 0.9889 Epoch 376/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9960 - val_loss: 0.0742 - val_accuracy: 0.9879 Epoch 377/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9965 - val_loss: 0.0725 - val_accuracy: 0.9889 Epoch 378/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9962 - val_loss: 0.0697 - val_accuracy: 0.9887 Epoch 379/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9964 - val_loss: 0.0703 - val_accuracy: 0.9888 Epoch 380/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0123 - accuracy: 0.9965 - val_loss: 0.0668 - val_accuracy: 0.9886 Epoch 381/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0137 - accuracy: 0.9962 - val_loss: 0.0641 - val_accuracy: 0.9883 Epoch 382/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0118 - accuracy: 0.9966 - val_loss: 0.0712 - val_accuracy: 0.9885 Epoch 383/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9960 - val_loss: 0.0726 - val_accuracy: 0.9883 Epoch 384/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0139 - accuracy: 0.9961 - val_loss: 0.0692 - val_accuracy: 0.9880 Epoch 385/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9964 - val_loss: 0.0640 - val_accuracy: 0.9885 Epoch 386/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0128 - accuracy: 0.9967 - val_loss: 0.0643 - val_accuracy: 0.9891 Epoch 387/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9967 - val_loss: 0.0628 - val_accuracy: 0.9899 Epoch 388/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0121 - accuracy: 0.9970 - val_loss: 0.0629 - val_accuracy: 0.9893 Epoch 389/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9964 - val_loss: 0.0663 - val_accuracy: 0.9888 Epoch 390/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0118 - accuracy: 0.9966 - val_loss: 0.0673 - val_accuracy: 0.9892 Epoch 391/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0116 - accuracy: 0.9966 - val_loss: 0.0680 - val_accuracy: 0.9899 Epoch 392/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9963 - val_loss: 0.0680 - val_accuracy: 0.9892 Epoch 393/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9966 - val_loss: 0.0686 - val_accuracy: 0.9894 Epoch 394/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0142 - accuracy: 0.9962 - val_loss: 0.0707 - val_accuracy: 0.9885 Epoch 395/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0115 - accuracy: 0.9970 - val_loss: 0.0737 - val_accuracy: 0.9883 Epoch 396/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0109 - accuracy: 0.9969 - val_loss: 0.0772 - val_accuracy: 0.9891 Epoch 397/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0115 - accuracy: 0.9969 - val_loss: 0.0709 - val_accuracy: 0.9883 Epoch 398/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0145 - accuracy: 0.9962 - val_loss: 0.0706 - val_accuracy: 0.9890 Epoch 399/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9963 - val_loss: 0.0724 - val_accuracy: 0.9885 Epoch 400/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0137 - accuracy: 0.9964 - val_loss: 0.0673 - val_accuracy: 0.9885 Epoch 401/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9966 - val_loss: 0.0686 - val_accuracy: 0.9890 Epoch 402/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9967 - val_loss: 0.0697 - val_accuracy: 0.9893 Epoch 403/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9969 - val_loss: 0.0649 - val_accuracy: 0.9897 Epoch 404/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0128 - accuracy: 0.9963 - val_loss: 0.0743 - val_accuracy: 0.9892 Epoch 405/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9962 - val_loss: 0.0630 - val_accuracy: 0.9893 Epoch 406/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0112 - accuracy: 0.9967 - val_loss: 0.0716 - val_accuracy: 0.9895 Epoch 407/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0130 - accuracy: 0.9965 - val_loss: 0.0675 - val_accuracy: 0.9893 Epoch 408/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9971 - val_loss: 0.0637 - val_accuracy: 0.9895 Epoch 409/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9967 - val_loss: 0.0692 - val_accuracy: 0.9888 Epoch 410/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9965 - val_loss: 0.0676 - val_accuracy: 0.9893 Epoch 411/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0121 - accuracy: 0.9967 - val_loss: 0.0673 - val_accuracy: 0.9895 Epoch 412/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0114 - accuracy: 0.9969 - val_loss: 0.0687 - val_accuracy: 0.9899 Epoch 413/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9964 - val_loss: 0.0654 - val_accuracy: 0.9898 Epoch 414/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9966 - val_loss: 0.0702 - val_accuracy: 0.9889 Epoch 415/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9967 - val_loss: 0.0707 - val_accuracy: 0.9888 Epoch 416/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0137 - accuracy: 0.9966 - val_loss: 0.0665 - val_accuracy: 0.9900 Epoch 417/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0115 - accuracy: 0.9969 - val_loss: 0.0654 - val_accuracy: 0.9894 Epoch 418/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9965 - val_loss: 0.0655 - val_accuracy: 0.9890 Epoch 419/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9965 - val_loss: 0.0663 - val_accuracy: 0.9896 Epoch 420/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9966 - val_loss: 0.0674 - val_accuracy: 0.9894 Epoch 421/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9963 - val_loss: 0.0622 - val_accuracy: 0.9898 Epoch 422/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9966 - val_loss: 0.0638 - val_accuracy: 0.9902 Epoch 423/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9967 - val_loss: 0.0644 - val_accuracy: 0.9905 Epoch 424/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9969 - val_loss: 0.0673 - val_accuracy: 0.9894 Epoch 425/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9967 - val_loss: 0.0692 - val_accuracy: 0.9895 Epoch 426/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9969 - val_loss: 0.0673 - val_accuracy: 0.9897 Epoch 427/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9965 - val_loss: 0.0676 - val_accuracy: 0.9881 Epoch 428/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0124 - accuracy: 0.9968 - val_loss: 0.0673 - val_accuracy: 0.9891 Epoch 429/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9965 - val_loss: 0.0677 - val_accuracy: 0.9897 Epoch 430/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9965 - val_loss: 0.0698 - val_accuracy: 0.9894 Epoch 431/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9962 - val_loss: 0.0700 - val_accuracy: 0.9886 Epoch 432/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9963 - val_loss: 0.0697 - val_accuracy: 0.9894 Epoch 433/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9965 - val_loss: 0.0639 - val_accuracy: 0.9892 Epoch 434/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9967 - val_loss: 0.0698 - val_accuracy: 0.9885 Epoch 435/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0124 - accuracy: 0.9969 - val_loss: 0.0664 - val_accuracy: 0.9888 Epoch 436/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9967 - val_loss: 0.0638 - val_accuracy: 0.9895 Epoch 437/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9963 - val_loss: 0.0651 - val_accuracy: 0.9888 Epoch 438/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0121 - accuracy: 0.9967 - val_loss: 0.0688 - val_accuracy: 0.9889 Epoch 439/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9965 - val_loss: 0.0670 - val_accuracy: 0.9892 Epoch 440/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9970 - val_loss: 0.0676 - val_accuracy: 0.9889 Epoch 441/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0130 - accuracy: 0.9964 - val_loss: 0.0686 - val_accuracy: 0.9884 Epoch 442/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9966 - val_loss: 0.0722 - val_accuracy: 0.9887 Epoch 443/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9968 - val_loss: 0.0719 - val_accuracy: 0.9884 Epoch 444/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9963 - val_loss: 0.0696 - val_accuracy: 0.9893 Epoch 445/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0114 - accuracy: 0.9969 - val_loss: 0.0705 - val_accuracy: 0.9894 Epoch 446/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9965 - val_loss: 0.0693 - val_accuracy: 0.9890 Epoch 447/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9964 - val_loss: 0.0774 - val_accuracy: 0.9889 Epoch 448/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9963 - val_loss: 0.0703 - val_accuracy: 0.9884 Epoch 449/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0143 - accuracy: 0.9963 - val_loss: 0.0689 - val_accuracy: 0.9900 Epoch 450/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9967 - val_loss: 0.0690 - val_accuracy: 0.9893 Epoch 451/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9971 - val_loss: 0.0677 - val_accuracy: 0.9894 Epoch 452/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0121 - accuracy: 0.9967 - val_loss: 0.0679 - val_accuracy: 0.9897 Epoch 453/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9968 - val_loss: 0.0706 - val_accuracy: 0.9892 Epoch 454/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9966 - val_loss: 0.0767 - val_accuracy: 0.9892 Epoch 455/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0139 - accuracy: 0.9962 - val_loss: 0.0712 - val_accuracy: 0.9879 Epoch 456/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9963 - val_loss: 0.0713 - val_accuracy: 0.9893 Epoch 457/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9966 - val_loss: 0.0709 - val_accuracy: 0.9889 Epoch 458/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0119 - accuracy: 0.9969 - val_loss: 0.0754 - val_accuracy: 0.9880 Epoch 459/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9966 - val_loss: 0.0764 - val_accuracy: 0.9883 Epoch 460/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0114 - accuracy: 0.9970 - val_loss: 0.0746 - val_accuracy: 0.9892 Epoch 461/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9968 - val_loss: 0.0752 - val_accuracy: 0.9885 Epoch 462/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9967 - val_loss: 0.0726 - val_accuracy: 0.9891 Epoch 463/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0113 - accuracy: 0.9969 - val_loss: 0.0775 - val_accuracy: 0.9896 Epoch 464/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9965 - val_loss: 0.0748 - val_accuracy: 0.9885 Epoch 465/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9966 - val_loss: 0.0763 - val_accuracy: 0.9883 Epoch 466/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9965 - val_loss: 0.0796 - val_accuracy: 0.9882 Epoch 467/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9967 - val_loss: 0.0743 - val_accuracy: 0.9888 Epoch 468/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0113 - accuracy: 0.9970 - val_loss: 0.0760 - val_accuracy: 0.9891 Epoch 469/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9968 - val_loss: 0.0768 - val_accuracy: 0.9890 Epoch 470/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9965 - val_loss: 0.0725 - val_accuracy: 0.9886 Epoch 471/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9966 - val_loss: 0.0720 - val_accuracy: 0.9888 Epoch 472/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0110 - accuracy: 0.9972 - val_loss: 0.0708 - val_accuracy: 0.9885 Epoch 473/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9968 - val_loss: 0.0699 - val_accuracy: 0.9886 Epoch 474/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0136 - accuracy: 0.9966 - val_loss: 0.0765 - val_accuracy: 0.9883 Epoch 475/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9968 - val_loss: 0.0735 - val_accuracy: 0.9879 Epoch 476/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9964 - val_loss: 0.0666 - val_accuracy: 0.9888 Epoch 477/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0123 - accuracy: 0.9969 - val_loss: 0.0683 - val_accuracy: 0.9892 Epoch 478/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0112 - accuracy: 0.9969 - val_loss: 0.0718 - val_accuracy: 0.9882 Epoch 479/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0114 - accuracy: 0.9969 - val_loss: 0.0705 - val_accuracy: 0.9886 Epoch 480/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9969 - val_loss: 0.0671 - val_accuracy: 0.9898 Epoch 481/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9964 - val_loss: 0.0778 - val_accuracy: 0.9883 Epoch 482/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0133 - accuracy: 0.9967 - val_loss: 0.0781 - val_accuracy: 0.9889 Epoch 483/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0120 - accuracy: 0.9969 - val_loss: 0.0768 - val_accuracy: 0.9882 Epoch 484/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0135 - accuracy: 0.9967 - val_loss: 0.0757 - val_accuracy: 0.9886 Epoch 485/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9962 - val_loss: 0.0683 - val_accuracy: 0.9882 Epoch 486/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0141 - accuracy: 0.9964 - val_loss: 0.0685 - val_accuracy: 0.9883 Epoch 487/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9966 - val_loss: 0.0697 - val_accuracy: 0.9884 Epoch 488/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0150 - accuracy: 0.9963 - val_loss: 0.0659 - val_accuracy: 0.9887 Epoch 489/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0112 - accuracy: 0.9972 - val_loss: 0.0682 - val_accuracy: 0.9892 Epoch 490/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9966 - val_loss: 0.0691 - val_accuracy: 0.9885 Epoch 491/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9966 - val_loss: 0.0658 - val_accuracy: 0.9889 Epoch 492/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0128 - accuracy: 0.9967 - val_loss: 0.0734 - val_accuracy: 0.9880 Epoch 493/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9964 - val_loss: 0.0746 - val_accuracy: 0.9883 Epoch 494/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0125 - accuracy: 0.9968 - val_loss: 0.0725 - val_accuracy: 0.9886 Epoch 495/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0132 - accuracy: 0.9968 - val_loss: 0.0679 - val_accuracy: 0.9892 Epoch 496/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0111 - accuracy: 0.9969 - val_loss: 0.0753 - val_accuracy: 0.9883 Epoch 497/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9968 - val_loss: 0.0785 - val_accuracy: 0.9885 Epoch 498/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0099 - accuracy: 0.9975 - val_loss: 0.0752 - val_accuracy: 0.9887 Epoch 499/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0121 - accuracy: 0.9963 - val_loss: 0.0722 - val_accuracy: 0.9890 Epoch 500/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0129 - accuracy: 0.9967 - val_loss: 0.0689 - val_accuracy: 0.9895
Activation function: ReLU; initialization: Kaiming He’s initializer; no dropout¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.HeNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten (Flatten) (None, 784) 0 _________________________________________________________________ dense (Dense) (None, 1024) 803840 _________________________________________________________________ dense_1 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_2 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_3 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_4 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_5 (Dense) (None, 10) 10250 ================================================================= Total params: 5,012,490 Trainable params: 5,012,490 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam()
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"])
history1 = model.fit(X_train, y_train, batch_size=128, epochs=500, validation_data=(X_test, y_test))
Epoch 1/500 469/469 [==============================] - 2s 5ms/step - loss: 0.2249 - accuracy: 0.9318 - val_loss: 0.1190 - val_accuracy: 0.9646 Epoch 2/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0928 - accuracy: 0.9729 - val_loss: 0.1528 - val_accuracy: 0.9613 Epoch 3/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0674 - accuracy: 0.9799 - val_loss: 0.0949 - val_accuracy: 0.9724 Epoch 4/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0508 - accuracy: 0.9849 - val_loss: 0.0775 - val_accuracy: 0.9781 Epoch 5/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0445 - accuracy: 0.9864 - val_loss: 0.0799 - val_accuracy: 0.9808 Epoch 6/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0368 - accuracy: 0.9893 - val_loss: 0.0847 - val_accuracy: 0.9801 Epoch 7/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0339 - accuracy: 0.9903 - val_loss: 0.0791 - val_accuracy: 0.9817 Epoch 8/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0321 - accuracy: 0.9909 - val_loss: 0.0875 - val_accuracy: 0.9787 Epoch 9/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0275 - accuracy: 0.9923 - val_loss: 0.0834 - val_accuracy: 0.9813 Epoch 10/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0269 - accuracy: 0.9924 - val_loss: 0.0858 - val_accuracy: 0.9794 Epoch 11/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0298 - accuracy: 0.9923 - val_loss: 0.0892 - val_accuracy: 0.9793 Epoch 12/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0203 - accuracy: 0.9950 - val_loss: 0.1187 - val_accuracy: 0.9794 Epoch 13/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0233 - accuracy: 0.9940 - val_loss: 0.0849 - val_accuracy: 0.9837 Epoch 14/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0222 - accuracy: 0.9946 - val_loss: 0.0934 - val_accuracy: 0.9841 Epoch 15/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0205 - accuracy: 0.9948 - val_loss: 0.1004 - val_accuracy: 0.9790 Epoch 16/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0231 - accuracy: 0.9943 - val_loss: 0.1270 - val_accuracy: 0.9799 Epoch 17/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0188 - accuracy: 0.9953 - val_loss: 0.1141 - val_accuracy: 0.9820 Epoch 18/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9971 - val_loss: 0.1153 - val_accuracy: 0.9827 Epoch 19/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0164 - accuracy: 0.9962 - val_loss: 0.1125 - val_accuracy: 0.9803 Epoch 20/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0188 - accuracy: 0.9957 - val_loss: 0.1235 - val_accuracy: 0.9816 Epoch 21/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0178 - accuracy: 0.9958 - val_loss: 0.0939 - val_accuracy: 0.9844 Epoch 22/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0166 - accuracy: 0.9962 - val_loss: 0.0941 - val_accuracy: 0.9846 Epoch 23/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0173 - accuracy: 0.9959 - val_loss: 0.1025 - val_accuracy: 0.9805 Epoch 24/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0129 - accuracy: 0.9969 - val_loss: 0.1467 - val_accuracy: 0.9783 Epoch 25/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0144 - accuracy: 0.9965 - val_loss: 0.1117 - val_accuracy: 0.9826 Epoch 26/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0134 - accuracy: 0.9971 - val_loss: 0.1213 - val_accuracy: 0.9828 Epoch 27/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0099 - accuracy: 0.9974 - val_loss: 0.1256 - val_accuracy: 0.9852 Epoch 28/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0112 - accuracy: 0.9972 - val_loss: 0.1439 - val_accuracy: 0.9838 Epoch 29/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0137 - accuracy: 0.9973 - val_loss: 0.1244 - val_accuracy: 0.9793 Epoch 30/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0122 - accuracy: 0.9976 - val_loss: 0.1576 - val_accuracy: 0.9805 Epoch 31/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0116 - accuracy: 0.9975 - val_loss: 0.1096 - val_accuracy: 0.9845 Epoch 32/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0148 - accuracy: 0.9974 - val_loss: 0.1108 - val_accuracy: 0.9846 Epoch 33/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0088 - accuracy: 0.9982 - val_loss: 0.1480 - val_accuracy: 0.9756 Epoch 34/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0166 - accuracy: 0.9973 - val_loss: 0.1542 - val_accuracy: 0.9796 Epoch 35/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0073 - accuracy: 0.9984 - val_loss: 0.1012 - val_accuracy: 0.9845 Epoch 36/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9979 - val_loss: 0.1098 - val_accuracy: 0.9831 Epoch 37/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9975 - val_loss: 0.1508 - val_accuracy: 0.9826 Epoch 38/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0131 - accuracy: 0.9972 - val_loss: 0.1388 - val_accuracy: 0.9833 Epoch 39/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0118 - accuracy: 0.9977 - val_loss: 0.1356 - val_accuracy: 0.9823 Epoch 40/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0061 - accuracy: 0.9986 - val_loss: 0.1383 - val_accuracy: 0.9823 Epoch 41/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0063 - accuracy: 0.9986 - val_loss: 0.1482 - val_accuracy: 0.9841 Epoch 42/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0104 - accuracy: 0.9977 - val_loss: 0.1621 - val_accuracy: 0.9829 Epoch 43/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0073 - accuracy: 0.9987 - val_loss: 0.1723 - val_accuracy: 0.9824 Epoch 44/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9981 - val_loss: 0.1467 - val_accuracy: 0.9832 Epoch 45/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0102 - accuracy: 0.9980 - val_loss: 0.1533 - val_accuracy: 0.9833 Epoch 46/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0063 - accuracy: 0.9985 - val_loss: 0.1853 - val_accuracy: 0.9837 Epoch 47/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0138 - accuracy: 0.9978 - val_loss: 0.1365 - val_accuracy: 0.9837 Epoch 48/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9983 - val_loss: 0.1824 - val_accuracy: 0.9810 Epoch 49/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0082 - accuracy: 0.9984 - val_loss: 0.1620 - val_accuracy: 0.9847 Epoch 50/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0074 - accuracy: 0.9988 - val_loss: 0.1419 - val_accuracy: 0.9820 Epoch 51/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0116 - accuracy: 0.9980 - val_loss: 0.1397 - val_accuracy: 0.9831 Epoch 52/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0079 - accuracy: 0.9983 - val_loss: 0.1095 - val_accuracy: 0.9848 Epoch 53/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0087 - accuracy: 0.9987 - val_loss: 0.1132 - val_accuracy: 0.9824 Epoch 54/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0036 - accuracy: 0.9993 - val_loss: 0.1660 - val_accuracy: 0.9851 Epoch 55/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9983 - val_loss: 0.1549 - val_accuracy: 0.9838 Epoch 56/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0063 - accuracy: 0.9989 - val_loss: 0.1878 - val_accuracy: 0.9818 Epoch 57/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0127 - accuracy: 0.9980 - val_loss: 0.1686 - val_accuracy: 0.9823 Epoch 58/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0126 - accuracy: 0.9979 - val_loss: 0.1467 - val_accuracy: 0.9832 Epoch 59/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0085 - accuracy: 0.9982 - val_loss: 0.1497 - val_accuracy: 0.9839 Epoch 60/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0056 - accuracy: 0.9990 - val_loss: 0.1738 - val_accuracy: 0.9834 Epoch 61/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0154 - accuracy: 0.9980 - val_loss: 0.1658 - val_accuracy: 0.9815 Epoch 62/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0085 - accuracy: 0.9981 - val_loss: 0.1635 - val_accuracy: 0.9834 Epoch 63/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0022 - accuracy: 0.9994 - val_loss: 0.1788 - val_accuracy: 0.9859 Epoch 64/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0014 - accuracy: 0.9997 - val_loss: 0.2588 - val_accuracy: 0.9833 Epoch 65/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0096 - accuracy: 0.9985 - val_loss: 0.1957 - val_accuracy: 0.9849 Epoch 66/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0100 - accuracy: 0.9986 - val_loss: 0.1739 - val_accuracy: 0.9776 Epoch 67/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0070 - accuracy: 0.9983 - val_loss: 0.1896 - val_accuracy: 0.9829 Epoch 68/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0078 - accuracy: 0.9984 - val_loss: 0.1490 - val_accuracy: 0.9849 Epoch 69/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0117 - accuracy: 0.9981 - val_loss: 0.1578 - val_accuracy: 0.9831 Epoch 70/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0094 - accuracy: 0.9984 - val_loss: 0.1507 - val_accuracy: 0.9865 Epoch 71/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0053 - accuracy: 0.9991 - val_loss: 0.1428 - val_accuracy: 0.9859 Epoch 72/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0050 - accuracy: 0.9990 - val_loss: 0.1631 - val_accuracy: 0.9863 Epoch 73/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0089 - accuracy: 0.9988 - val_loss: 0.1880 - val_accuracy: 0.9814 Epoch 74/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0141 - accuracy: 0.9977 - val_loss: 0.1280 - val_accuracy: 0.9839 Epoch 75/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0013 - accuracy: 0.9998 - val_loss: 0.1745 - val_accuracy: 0.9871 Epoch 76/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0018 - accuracy: 0.9997 - val_loss: 0.1707 - val_accuracy: 0.9867 Epoch 77/500 469/469 [==============================] - 2s 4ms/step - loss: 6.0835e-06 - accuracy: 1.0000 - val_loss: 0.1884 - val_accuracy: 0.9864 Epoch 78/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1267e-06 - accuracy: 1.0000 - val_loss: 0.1922 - val_accuracy: 0.9866 Epoch 79/500 469/469 [==============================] - 2s 4ms/step - loss: 6.9750e-07 - accuracy: 1.0000 - val_loss: 0.1960 - val_accuracy: 0.9866 Epoch 80/500 469/469 [==============================] - 2s 4ms/step - loss: 4.4371e-07 - accuracy: 1.0000 - val_loss: 0.2001 - val_accuracy: 0.9866 Epoch 81/500 469/469 [==============================] - 2s 4ms/step - loss: 2.8882e-07 - accuracy: 1.0000 - val_loss: 0.2040 - val_accuracy: 0.9867 Epoch 82/500 469/469 [==============================] - 2s 4ms/step - loss: 1.8003e-07 - accuracy: 1.0000 - val_loss: 0.2087 - val_accuracy: 0.9868 Epoch 83/500 469/469 [==============================] - 2s 4ms/step - loss: 9.0416e-08 - accuracy: 1.0000 - val_loss: 0.2142 - val_accuracy: 0.9868 Epoch 84/500 469/469 [==============================] - 2s 4ms/step - loss: 5.1665e-08 - accuracy: 1.0000 - val_loss: 0.2181 - val_accuracy: 0.9868 Epoch 85/500 469/469 [==============================] - 2s 4ms/step - loss: 3.5470e-08 - accuracy: 1.0000 - val_loss: 0.2219 - val_accuracy: 0.9868 Epoch 86/500 469/469 [==============================] - 2s 4ms/step - loss: 2.5689e-08 - accuracy: 1.0000 - val_loss: 0.2253 - val_accuracy: 0.9868 Epoch 87/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9099e-08 - accuracy: 1.0000 - val_loss: 0.2286 - val_accuracy: 0.9869 Epoch 88/500 469/469 [==============================] - 2s 4ms/step - loss: 1.4506e-08 - accuracy: 1.0000 - val_loss: 0.2317 - val_accuracy: 0.9869 Epoch 89/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1176e-08 - accuracy: 1.0000 - val_loss: 0.2348 - val_accuracy: 0.9870 Epoch 90/500 469/469 [==============================] - 2s 4ms/step - loss: 8.5393e-09 - accuracy: 1.0000 - val_loss: 0.2376 - val_accuracy: 0.9870 Epoch 91/500 469/469 [==============================] - 2s 4ms/step - loss: 6.5863e-09 - accuracy: 1.0000 - val_loss: 0.2405 - val_accuracy: 0.9870 Epoch 92/500 469/469 [==============================] - 2s 4ms/step - loss: 5.1300e-09 - accuracy: 1.0000 - val_loss: 0.2432 - val_accuracy: 0.9871 Epoch 93/500 469/469 [==============================] - 2s 4ms/step - loss: 4.0392e-09 - accuracy: 1.0000 - val_loss: 0.2458 - val_accuracy: 0.9871 Epoch 94/500 469/469 [==============================] - 2s 4ms/step - loss: 3.2266e-09 - accuracy: 1.0000 - val_loss: 0.2483 - val_accuracy: 0.9871 Epoch 95/500 469/469 [==============================] - 2s 4ms/step - loss: 2.5988e-09 - accuracy: 1.0000 - val_loss: 0.2507 - val_accuracy: 0.9870 Epoch 96/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0742e-09 - accuracy: 1.0000 - val_loss: 0.2530 - val_accuracy: 0.9870 Epoch 97/500 469/469 [==============================] - 2s 4ms/step - loss: 1.7007e-09 - accuracy: 1.0000 - val_loss: 0.2551 - val_accuracy: 0.9870 Epoch 98/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3789e-09 - accuracy: 1.0000 - val_loss: 0.2574 - val_accuracy: 0.9871 Epoch 99/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1444e-09 - accuracy: 1.0000 - val_loss: 0.2594 - val_accuracy: 0.9871 Epoch 100/500 469/469 [==============================] - 2s 4ms/step - loss: 9.4374e-10 - accuracy: 1.0000 - val_loss: 0.2614 - val_accuracy: 0.9871 Epoch 101/500 469/469 [==============================] - 2s 4ms/step - loss: 7.7486e-10 - accuracy: 1.0000 - val_loss: 0.2633 - val_accuracy: 0.9871 Epoch 102/500 469/469 [==============================] - 2s 4ms/step - loss: 6.2784e-10 - accuracy: 1.0000 - val_loss: 0.2652 - val_accuracy: 0.9871 Epoch 103/500 469/469 [==============================] - 2s 4ms/step - loss: 5.2452e-10 - accuracy: 1.0000 - val_loss: 0.2670 - val_accuracy: 0.9871 Epoch 104/500 469/469 [==============================] - 2s 4ms/step - loss: 4.3313e-10 - accuracy: 1.0000 - val_loss: 0.2688 - val_accuracy: 0.9871 Epoch 105/500 469/469 [==============================] - 2s 4ms/step - loss: 3.6160e-10 - accuracy: 1.0000 - val_loss: 0.2705 - val_accuracy: 0.9871 Epoch 106/500 469/469 [==============================] - 2s 4ms/step - loss: 2.9405e-10 - accuracy: 1.0000 - val_loss: 0.2722 - val_accuracy: 0.9871 Epoch 107/500 469/469 [==============================] - 2s 4ms/step - loss: 2.4835e-10 - accuracy: 1.0000 - val_loss: 0.2738 - val_accuracy: 0.9871 Epoch 108/500 469/469 [==============================] - 2s 4ms/step - loss: 2.0663e-10 - accuracy: 1.0000 - val_loss: 0.2754 - val_accuracy: 0.9871 Epoch 109/500 469/469 [==============================] - 2s 4ms/step - loss: 1.6292e-10 - accuracy: 1.0000 - val_loss: 0.2770 - val_accuracy: 0.9871 Epoch 110/500 469/469 [==============================] - 2s 4ms/step - loss: 1.2914e-10 - accuracy: 1.0000 - val_loss: 0.2785 - val_accuracy: 0.9871 Epoch 111/500 469/469 [==============================] - 2s 4ms/step - loss: 1.1126e-10 - accuracy: 1.0000 - val_loss: 0.2800 - val_accuracy: 0.9871 Epoch 112/500 469/469 [==============================] - 2s 4ms/step - loss: 8.5433e-11 - accuracy: 1.0000 - val_loss: 0.2815 - val_accuracy: 0.9871 Epoch 113/500 469/469 [==============================] - 2s 4ms/step - loss: 7.3512e-11 - accuracy: 1.0000 - val_loss: 0.2829 - val_accuracy: 0.9871 Epoch 114/500 469/469 [==============================] - 2s 4ms/step - loss: 5.7618e-11 - accuracy: 1.0000 - val_loss: 0.2843 - val_accuracy: 0.9871 Epoch 115/500 469/469 [==============================] - 2s 4ms/step - loss: 4.7684e-11 - accuracy: 1.0000 - val_loss: 0.2856 - val_accuracy: 0.9871 Epoch 116/500 469/469 [==============================] - 2s 4ms/step - loss: 3.5763e-11 - accuracy: 1.0000 - val_loss: 0.2869 - val_accuracy: 0.9871 Epoch 117/500 469/469 [==============================] - 2s 4ms/step - loss: 2.9802e-11 - accuracy: 1.0000 - val_loss: 0.2882 - val_accuracy: 0.9872 Epoch 118/500 469/469 [==============================] - 2s 4ms/step - loss: 2.7815e-11 - accuracy: 1.0000 - val_loss: 0.2894 - val_accuracy: 0.9872 Epoch 119/500 469/469 [==============================] - 2s 4ms/step - loss: 2.1855e-11 - accuracy: 1.0000 - val_loss: 0.2906 - val_accuracy: 0.9872 Epoch 120/500 469/469 [==============================] - 2s 4ms/step - loss: 1.3908e-11 - accuracy: 1.0000 - val_loss: 0.2917 - val_accuracy: 0.9872 Epoch 121/500 469/469 [==============================] - 2s 4ms/step - loss: 7.9473e-12 - accuracy: 1.0000 - val_loss: 0.2928 - val_accuracy: 0.9872 Epoch 122/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2938 - val_accuracy: 0.9872 Epoch 123/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2947 - val_accuracy: 0.9872 Epoch 124/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2957 - val_accuracy: 0.9872 Epoch 125/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2966 - val_accuracy: 0.9872 Epoch 126/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2974 - val_accuracy: 0.9872 Epoch 127/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2983 - val_accuracy: 0.9872 Epoch 128/500 469/469 [==============================] - 2s 4ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2991 - val_accuracy: 0.9872 Epoch 129/500 469/469 [==============================] - 2s 5ms/step - loss: 3.9736e-12 - accuracy: 1.0000 - val_loss: 0.2998 - val_accuracy: 0.9872 Epoch 130/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3005 - val_accuracy: 0.9871 Epoch 131/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3011 - val_accuracy: 0.9871 Epoch 132/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3017 - val_accuracy: 0.9871 Epoch 133/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3023 - val_accuracy: 0.9871 Epoch 134/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3029 - val_accuracy: 0.9871 Epoch 135/500 469/469 [==============================] - 2s 4ms/step - loss: 1.9868e-12 - accuracy: 1.0000 - val_loss: 0.3035 - val_accuracy: 0.9871 Epoch 136/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3040 - val_accuracy: 0.9871 Epoch 137/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3045 - val_accuracy: 0.9871 Epoch 138/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3049 - val_accuracy: 0.9871 Epoch 139/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3054 - val_accuracy: 0.9871 Epoch 140/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3058 - val_accuracy: 0.9871 Epoch 141/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3063 - val_accuracy: 0.9871 Epoch 142/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3066 - val_accuracy: 0.9871 Epoch 143/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3070 - val_accuracy: 0.9871 Epoch 144/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3074 - val_accuracy: 0.9871 Epoch 145/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3078 - val_accuracy: 0.9871 Epoch 146/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3081 - val_accuracy: 0.9871 Epoch 147/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3085 - val_accuracy: 0.9871 Epoch 148/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3088 - val_accuracy: 0.9871 Epoch 149/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3091 - val_accuracy: 0.9871 Epoch 150/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3094 - val_accuracy: 0.9871 Epoch 151/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3097 - val_accuracy: 0.9871 Epoch 152/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3100 - val_accuracy: 0.9871 Epoch 153/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3103 - val_accuracy: 0.9871 Epoch 154/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3106 - val_accuracy: 0.9871 Epoch 155/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3108 - val_accuracy: 0.9871 Epoch 156/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3111 - val_accuracy: 0.9871 Epoch 157/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3114 - val_accuracy: 0.9871 Epoch 158/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3116 - val_accuracy: 0.9871 Epoch 159/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3118 - val_accuracy: 0.9871 Epoch 160/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3121 - val_accuracy: 0.9871 Epoch 161/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3123 - val_accuracy: 0.9871 Epoch 162/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3126 - val_accuracy: 0.9871 Epoch 163/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3128 - val_accuracy: 0.9871 Epoch 164/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3130 - val_accuracy: 0.9871 Epoch 165/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3132 - val_accuracy: 0.9871 Epoch 166/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3134 - val_accuracy: 0.9871 Epoch 167/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3136 - val_accuracy: 0.9871 Epoch 168/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3138 - val_accuracy: 0.9871 Epoch 169/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3140 - val_accuracy: 0.9871 Epoch 170/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3142 - val_accuracy: 0.9871 Epoch 171/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3144 - val_accuracy: 0.9871 Epoch 172/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3146 - val_accuracy: 0.9871 Epoch 173/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3148 - val_accuracy: 0.9871 Epoch 174/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3150 - val_accuracy: 0.9871 Epoch 175/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3152 - val_accuracy: 0.9871 Epoch 176/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3153 - val_accuracy: 0.9871 Epoch 177/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3155 - val_accuracy: 0.9871 Epoch 178/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3157 - val_accuracy: 0.9871 Epoch 179/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3158 - val_accuracy: 0.9871 Epoch 180/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3160 - val_accuracy: 0.9871 Epoch 181/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3162 - val_accuracy: 0.9871 Epoch 182/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3163 - val_accuracy: 0.9871 Epoch 183/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3165 - val_accuracy: 0.9871 Epoch 184/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3166 - val_accuracy: 0.9871 Epoch 185/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3168 - val_accuracy: 0.9871 Epoch 186/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3169 - val_accuracy: 0.9871 Epoch 187/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3171 - val_accuracy: 0.9871 Epoch 188/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3172 - val_accuracy: 0.9871 Epoch 189/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3174 - val_accuracy: 0.9871 Epoch 190/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3175 - val_accuracy: 0.9871 Epoch 191/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3177 - val_accuracy: 0.9871 Epoch 192/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3178 - val_accuracy: 0.9871 Epoch 193/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3179 - val_accuracy: 0.9871 Epoch 194/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3181 - val_accuracy: 0.9871 Epoch 195/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3182 - val_accuracy: 0.9871 Epoch 196/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3183 - val_accuracy: 0.9871 Epoch 197/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3185 - val_accuracy: 0.9871 Epoch 198/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3186 - val_accuracy: 0.9871 Epoch 199/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3187 - val_accuracy: 0.9871 Epoch 200/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3188 - val_accuracy: 0.9871 Epoch 201/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3190 - val_accuracy: 0.9871 Epoch 202/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3191 - val_accuracy: 0.9871 Epoch 203/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3192 - val_accuracy: 0.9871 Epoch 204/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3193 - val_accuracy: 0.9871 Epoch 205/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3194 - val_accuracy: 0.9871 Epoch 206/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3195 - val_accuracy: 0.9871 Epoch 207/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3197 - val_accuracy: 0.9871 Epoch 208/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3198 - val_accuracy: 0.9871 Epoch 209/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3199 - val_accuracy: 0.9871 Epoch 210/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3200 - val_accuracy: 0.9871 Epoch 211/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3201 - val_accuracy: 0.9871 Epoch 212/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3202 - val_accuracy: 0.9871 Epoch 213/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3203 - val_accuracy: 0.9871 Epoch 214/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3204 - val_accuracy: 0.9871 Epoch 215/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3205 - val_accuracy: 0.9871 Epoch 216/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3206 - val_accuracy: 0.9871 Epoch 217/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3207 - val_accuracy: 0.9871 Epoch 218/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3208 - val_accuracy: 0.9871 Epoch 219/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3209 - val_accuracy: 0.9871 Epoch 220/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3210 - val_accuracy: 0.9871 Epoch 221/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3211 - val_accuracy: 0.9871 Epoch 222/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3212 - val_accuracy: 0.9871 Epoch 223/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3213 - val_accuracy: 0.9871 Epoch 224/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3214 - val_accuracy: 0.9871 Epoch 225/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3215 - val_accuracy: 0.9871 Epoch 226/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3216 - val_accuracy: 0.9871 Epoch 227/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3217 - val_accuracy: 0.9871 Epoch 228/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3218 - val_accuracy: 0.9871 Epoch 229/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3219 - val_accuracy: 0.9871 Epoch 230/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3220 - val_accuracy: 0.9871 Epoch 231/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3221 - val_accuracy: 0.9871 Epoch 232/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3222 - val_accuracy: 0.9871 Epoch 233/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3222 - val_accuracy: 0.9872 Epoch 234/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3223 - val_accuracy: 0.9872 Epoch 235/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3224 - val_accuracy: 0.9872 Epoch 236/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3225 - val_accuracy: 0.9872 Epoch 237/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3226 - val_accuracy: 0.9872 Epoch 238/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3227 - val_accuracy: 0.9872 Epoch 239/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3227 - val_accuracy: 0.9872 Epoch 240/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3228 - val_accuracy: 0.9872 Epoch 241/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3229 - val_accuracy: 0.9872 Epoch 242/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3230 - val_accuracy: 0.9872 Epoch 243/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3231 - val_accuracy: 0.9872 Epoch 244/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3232 - val_accuracy: 0.9872 Epoch 245/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3232 - val_accuracy: 0.9872 Epoch 246/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3233 - val_accuracy: 0.9872 Epoch 247/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3234 - val_accuracy: 0.9872 Epoch 248/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3235 - val_accuracy: 0.9872 Epoch 249/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3235 - val_accuracy: 0.9872 Epoch 250/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3236 - val_accuracy: 0.9872 Epoch 251/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3237 - val_accuracy: 0.9872 Epoch 252/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3238 - val_accuracy: 0.9872 Epoch 253/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3238 - val_accuracy: 0.9872 Epoch 254/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3239 - val_accuracy: 0.9872 Epoch 255/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3240 - val_accuracy: 0.9872 Epoch 256/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3241 - val_accuracy: 0.9872 Epoch 257/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3241 - val_accuracy: 0.9872 Epoch 258/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3242 - val_accuracy: 0.9872 Epoch 259/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3243 - val_accuracy: 0.9873 Epoch 260/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3243 - val_accuracy: 0.9873 Epoch 261/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3244 - val_accuracy: 0.9873 Epoch 262/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3245 - val_accuracy: 0.9873 Epoch 263/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3246 - val_accuracy: 0.9873 Epoch 264/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3246 - val_accuracy: 0.9873 Epoch 265/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3247 - val_accuracy: 0.9873 Epoch 266/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3248 - val_accuracy: 0.9873 Epoch 267/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3248 - val_accuracy: 0.9873 Epoch 268/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3249 - val_accuracy: 0.9873 Epoch 269/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3250 - val_accuracy: 0.9873 Epoch 270/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3250 - val_accuracy: 0.9873 Epoch 271/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3251 - val_accuracy: 0.9873 Epoch 272/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3251 - val_accuracy: 0.9873 Epoch 273/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3252 - val_accuracy: 0.9873 Epoch 274/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3253 - val_accuracy: 0.9873 Epoch 275/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3253 - val_accuracy: 0.9873 Epoch 276/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3254 - val_accuracy: 0.9873 Epoch 277/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3255 - val_accuracy: 0.9873 Epoch 278/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3255 - val_accuracy: 0.9873 Epoch 279/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3256 - val_accuracy: 0.9873 Epoch 280/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3256 - val_accuracy: 0.9873 Epoch 281/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3257 - val_accuracy: 0.9873 Epoch 282/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3258 - val_accuracy: 0.9873 Epoch 283/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3258 - val_accuracy: 0.9873 Epoch 284/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3259 - val_accuracy: 0.9873 Epoch 285/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3259 - val_accuracy: 0.9873 Epoch 286/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3260 - val_accuracy: 0.9873 Epoch 287/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3261 - val_accuracy: 0.9873 Epoch 288/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3261 - val_accuracy: 0.9873 Epoch 289/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3262 - val_accuracy: 0.9873 Epoch 290/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3262 - val_accuracy: 0.9873 Epoch 291/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3263 - val_accuracy: 0.9873 Epoch 292/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3264 - val_accuracy: 0.9873 Epoch 293/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3264 - val_accuracy: 0.9873 Epoch 294/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3265 - val_accuracy: 0.9873 Epoch 295/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3265 - val_accuracy: 0.9873 Epoch 296/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3266 - val_accuracy: 0.9873 Epoch 297/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3266 - val_accuracy: 0.9873 Epoch 298/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3267 - val_accuracy: 0.9873 Epoch 299/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3267 - val_accuracy: 0.9873 Epoch 300/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3268 - val_accuracy: 0.9873 Epoch 301/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3269 - val_accuracy: 0.9873 Epoch 302/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3269 - val_accuracy: 0.9873 Epoch 303/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3270 - val_accuracy: 0.9873 Epoch 304/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3270 - val_accuracy: 0.9873 Epoch 305/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3271 - val_accuracy: 0.9873 Epoch 306/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3271 - val_accuracy: 0.9873 Epoch 307/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3272 - val_accuracy: 0.9873 Epoch 308/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3272 - val_accuracy: 0.9873 Epoch 309/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3273 - val_accuracy: 0.9873 Epoch 310/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3273 - val_accuracy: 0.9873 Epoch 311/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3274 - val_accuracy: 0.9873 Epoch 312/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3274 - val_accuracy: 0.9873 Epoch 313/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3275 - val_accuracy: 0.9873 Epoch 314/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3275 - val_accuracy: 0.9873 Epoch 315/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3276 - val_accuracy: 0.9873 Epoch 316/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3276 - val_accuracy: 0.9873 Epoch 317/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3277 - val_accuracy: 0.9873 Epoch 318/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3277 - val_accuracy: 0.9873 Epoch 319/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3278 - val_accuracy: 0.9873 Epoch 320/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3278 - val_accuracy: 0.9873 Epoch 321/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3279 - val_accuracy: 0.9873 Epoch 322/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3279 - val_accuracy: 0.9873 Epoch 323/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3280 - val_accuracy: 0.9873 Epoch 324/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3280 - val_accuracy: 0.9873 Epoch 325/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3281 - val_accuracy: 0.9873 Epoch 326/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3281 - val_accuracy: 0.9873 Epoch 327/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3281 - val_accuracy: 0.9873 Epoch 328/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3282 - val_accuracy: 0.9873 Epoch 329/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3282 - val_accuracy: 0.9873 Epoch 330/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3283 - val_accuracy: 0.9873 Epoch 331/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3283 - val_accuracy: 0.9873 Epoch 332/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3284 - val_accuracy: 0.9873 Epoch 333/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3284 - val_accuracy: 0.9873 Epoch 334/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3285 - val_accuracy: 0.9873 Epoch 335/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3285 - val_accuracy: 0.9873 Epoch 336/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3286 - val_accuracy: 0.9873 Epoch 337/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3286 - val_accuracy: 0.9873 Epoch 338/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3287 - val_accuracy: 0.9873 Epoch 339/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3287 - val_accuracy: 0.9873 Epoch 340/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3287 - val_accuracy: 0.9873 Epoch 341/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3288 - val_accuracy: 0.9873 Epoch 342/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3288 - val_accuracy: 0.9873 Epoch 343/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3289 - val_accuracy: 0.9873 Epoch 344/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3289 - val_accuracy: 0.9873 Epoch 345/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3290 - val_accuracy: 0.9873 Epoch 346/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3290 - val_accuracy: 0.9873 Epoch 347/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3290 - val_accuracy: 0.9873 Epoch 348/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3291 - val_accuracy: 0.9873 Epoch 349/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3291 - val_accuracy: 0.9873 Epoch 350/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3292 - val_accuracy: 0.9873 Epoch 351/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3292 - val_accuracy: 0.9873 Epoch 352/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3292 - val_accuracy: 0.9873 Epoch 353/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3293 - val_accuracy: 0.9873 Epoch 354/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3293 - val_accuracy: 0.9873 Epoch 355/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3294 - val_accuracy: 0.9873 Epoch 356/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3294 - val_accuracy: 0.9873 Epoch 357/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3295 - val_accuracy: 0.9873 Epoch 358/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3295 - val_accuracy: 0.9873 Epoch 359/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3295 - val_accuracy: 0.9873 Epoch 360/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3296 - val_accuracy: 0.9873 Epoch 361/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3296 - val_accuracy: 0.9873 Epoch 362/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3297 - val_accuracy: 0.9873 Epoch 363/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3297 - val_accuracy: 0.9873 Epoch 364/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3297 - val_accuracy: 0.9873 Epoch 365/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3298 - val_accuracy: 0.9873 Epoch 366/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3298 - val_accuracy: 0.9873 Epoch 367/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3299 - val_accuracy: 0.9873 Epoch 368/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3299 - val_accuracy: 0.9873 Epoch 369/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3299 - val_accuracy: 0.9873 Epoch 370/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3300 - val_accuracy: 0.9874 Epoch 371/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3300 - val_accuracy: 0.9874 Epoch 372/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3301 - val_accuracy: 0.9874 Epoch 373/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3301 - val_accuracy: 0.9874 Epoch 374/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3301 - val_accuracy: 0.9874 Epoch 375/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3302 - val_accuracy: 0.9874 Epoch 376/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3302 - val_accuracy: 0.9874 Epoch 377/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3302 - val_accuracy: 0.9874 Epoch 378/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3303 - val_accuracy: 0.9874 Epoch 379/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3303 - val_accuracy: 0.9874 Epoch 380/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3304 - val_accuracy: 0.9874 Epoch 381/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3304 - val_accuracy: 0.9874 Epoch 382/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3304 - val_accuracy: 0.9874 Epoch 383/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3305 - val_accuracy: 0.9874 Epoch 384/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3305 - val_accuracy: 0.9874 Epoch 385/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3305 - val_accuracy: 0.9874 Epoch 386/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3306 - val_accuracy: 0.9874 Epoch 387/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3306 - val_accuracy: 0.9874 Epoch 388/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3306 - val_accuracy: 0.9874 Epoch 389/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3307 - val_accuracy: 0.9874 Epoch 390/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3307 - val_accuracy: 0.9874 Epoch 391/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3308 - val_accuracy: 0.9874 Epoch 392/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3308 - val_accuracy: 0.9874 Epoch 393/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3308 - val_accuracy: 0.9874 Epoch 394/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3309 - val_accuracy: 0.9874 Epoch 395/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3309 - val_accuracy: 0.9874 Epoch 396/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3309 - val_accuracy: 0.9874 Epoch 397/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3310 - val_accuracy: 0.9874 Epoch 398/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3310 - val_accuracy: 0.9874 Epoch 399/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3310 - val_accuracy: 0.9874 Epoch 400/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3311 - val_accuracy: 0.9874 Epoch 401/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3311 - val_accuracy: 0.9874 Epoch 402/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3311 - val_accuracy: 0.9874 Epoch 403/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3312 - val_accuracy: 0.9874 Epoch 404/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3312 - val_accuracy: 0.9874 Epoch 405/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3312 - val_accuracy: 0.9874 Epoch 406/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3313 - val_accuracy: 0.9874 Epoch 407/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3313 - val_accuracy: 0.9874 Epoch 408/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3313 - val_accuracy: 0.9874 Epoch 409/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3314 - val_accuracy: 0.9874 Epoch 410/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3314 - val_accuracy: 0.9874 Epoch 411/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3314 - val_accuracy: 0.9874 Epoch 412/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3315 - val_accuracy: 0.9874 Epoch 413/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3315 - val_accuracy: 0.9874 Epoch 414/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3315 - val_accuracy: 0.9874 Epoch 415/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3316 - val_accuracy: 0.9874 Epoch 416/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3316 - val_accuracy: 0.9874 Epoch 417/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3316 - val_accuracy: 0.9874 Epoch 418/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3317 - val_accuracy: 0.9874 Epoch 419/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3317 - val_accuracy: 0.9874 Epoch 420/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3317 - val_accuracy: 0.9874 Epoch 421/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3318 - val_accuracy: 0.9874 Epoch 422/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3318 - val_accuracy: 0.9874 Epoch 423/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3318 - val_accuracy: 0.9874 Epoch 424/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3319 - val_accuracy: 0.9874 Epoch 425/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3319 - val_accuracy: 0.9874 Epoch 426/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3319 - val_accuracy: 0.9874 Epoch 427/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3320 - val_accuracy: 0.9874 Epoch 428/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3320 - val_accuracy: 0.9874 Epoch 429/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3320 - val_accuracy: 0.9874 Epoch 430/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3321 - val_accuracy: 0.9874 Epoch 431/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3321 - val_accuracy: 0.9874 Epoch 432/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3321 - val_accuracy: 0.9874 Epoch 433/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3321 - val_accuracy: 0.9874 Epoch 434/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3322 - val_accuracy: 0.9874 Epoch 435/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3322 - val_accuracy: 0.9874 Epoch 436/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3322 - val_accuracy: 0.9874 Epoch 437/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3323 - val_accuracy: 0.9874 Epoch 438/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3323 - val_accuracy: 0.9874 Epoch 439/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3323 - val_accuracy: 0.9874 Epoch 440/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3324 - val_accuracy: 0.9874 Epoch 441/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3324 - val_accuracy: 0.9874 Epoch 442/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3324 - val_accuracy: 0.9874 Epoch 443/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3325 - val_accuracy: 0.9874 Epoch 444/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3325 - val_accuracy: 0.9874 Epoch 445/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3325 - val_accuracy: 0.9874 Epoch 446/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3325 - val_accuracy: 0.9874 Epoch 447/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3326 - val_accuracy: 0.9874 Epoch 448/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3326 - val_accuracy: 0.9874 Epoch 449/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3326 - val_accuracy: 0.9874 Epoch 450/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3327 - val_accuracy: 0.9874 Epoch 451/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3327 - val_accuracy: 0.9874 Epoch 452/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3327 - val_accuracy: 0.9874 Epoch 453/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3327 - val_accuracy: 0.9874 Epoch 454/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3328 - val_accuracy: 0.9874 Epoch 455/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3328 - val_accuracy: 0.9874 Epoch 456/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3328 - val_accuracy: 0.9874 Epoch 457/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3329 - val_accuracy: 0.9874 Epoch 458/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3329 - val_accuracy: 0.9874 Epoch 459/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3329 - val_accuracy: 0.9874 Epoch 460/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3329 - val_accuracy: 0.9874 Epoch 461/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3330 - val_accuracy: 0.9874 Epoch 462/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3330 - val_accuracy: 0.9874 Epoch 463/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3330 - val_accuracy: 0.9874 Epoch 464/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3331 - val_accuracy: 0.9874 Epoch 465/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3331 - val_accuracy: 0.9874 Epoch 466/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3331 - val_accuracy: 0.9874 Epoch 467/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3331 - val_accuracy: 0.9874 Epoch 468/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3332 - val_accuracy: 0.9874 Epoch 469/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3332 - val_accuracy: 0.9874 Epoch 470/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3332 - val_accuracy: 0.9874 Epoch 471/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3332 - val_accuracy: 0.9874 Epoch 472/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3333 - val_accuracy: 0.9874 Epoch 473/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3333 - val_accuracy: 0.9874 Epoch 474/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3333 - val_accuracy: 0.9874 Epoch 475/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3334 - val_accuracy: 0.9874 Epoch 476/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3334 - val_accuracy: 0.9874 Epoch 477/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3334 - val_accuracy: 0.9874 Epoch 478/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3334 - val_accuracy: 0.9874 Epoch 479/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3335 - val_accuracy: 0.9874 Epoch 480/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3335 - val_accuracy: 0.9874 Epoch 481/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3335 - val_accuracy: 0.9874 Epoch 482/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3335 - val_accuracy: 0.9874 Epoch 483/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3336 - val_accuracy: 0.9874 Epoch 484/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3336 - val_accuracy: 0.9874 Epoch 485/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3336 - val_accuracy: 0.9874 Epoch 486/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3336 - val_accuracy: 0.9874 Epoch 487/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3337 - val_accuracy: 0.9874 Epoch 488/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3337 - val_accuracy: 0.9874 Epoch 489/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3337 - val_accuracy: 0.9874 Epoch 490/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3338 - val_accuracy: 0.9874 Epoch 491/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3338 - val_accuracy: 0.9874 Epoch 492/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3338 - val_accuracy: 0.9874 Epoch 493/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3338 - val_accuracy: 0.9874 Epoch 494/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3339 - val_accuracy: 0.9874 Epoch 495/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3339 - val_accuracy: 0.9874 Epoch 496/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3339 - val_accuracy: 0.9874 Epoch 497/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3339 - val_accuracy: 0.9874 Epoch 498/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3340 - val_accuracy: 0.9874 Epoch 499/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3340 - val_accuracy: 0.9874 Epoch 500/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.3340 - val_accuracy: 0.9874
Activation function: ReLU; initialization: Kaiming He’s initializer; with dropout rate: 0.2 for the first layer and 0.5 for the other hidden layers¶
In [ ]:
shape = (28, 28) # Define shape of input for Keras model
init = tf.keras.initializers.HeNormal(seed=None)
model = keras.Sequential(
[
tf.keras.layers.Input(shape=shape),
tf.keras.layers.Flatten(),
tf.keras.layers.Dropout(0.2),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dropout(0.5),
tf.keras.layers.Dense(1024,activation='relu',kernel_initializer=init),
tf.keras.layers.Dense(10, activation='softmax', kernel_initializer=init)
]
)
model.summary()
Model: "sequential_1" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= flatten_1 (Flatten) (None, 784) 0 _________________________________________________________________ dropout (Dropout) (None, 784) 0 _________________________________________________________________ dense_6 (Dense) (None, 1024) 803840 _________________________________________________________________ dropout_1 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_7 (Dense) (None, 1024) 1049600 _________________________________________________________________ dropout_2 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_8 (Dense) (None, 1024) 1049600 _________________________________________________________________ dropout_3 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_9 (Dense) (None, 1024) 1049600 _________________________________________________________________ dropout_4 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_10 (Dense) (None, 1024) 1049600 _________________________________________________________________ dense_11 (Dense) (None, 10) 10250 ================================================================= Total params: 5,012,490 Trainable params: 5,012,490 Non-trainable params: 0 _________________________________________________________________
In [ ]:
opt = keras.optimizers.Adam()
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"])
historyd1 = model.fit(X_train, y_train, batch_size=128, epochs=500, validation_data=(X_test, y_test))
Epoch 1/500 469/469 [==============================] - 2s 5ms/step - loss: 0.6479 - accuracy: 0.7908 - val_loss: 0.1693 - val_accuracy: 0.9496 Epoch 2/500 469/469 [==============================] - 2s 4ms/step - loss: 0.2751 - accuracy: 0.9209 - val_loss: 0.1315 - val_accuracy: 0.9623 Epoch 3/500 469/469 [==============================] - 2s 4ms/step - loss: 0.2158 - accuracy: 0.9384 - val_loss: 0.1165 - val_accuracy: 0.9685 Epoch 4/500 469/469 [==============================] - 2s 5ms/step - loss: 0.1927 - accuracy: 0.9469 - val_loss: 0.0935 - val_accuracy: 0.9730 Epoch 5/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1760 - accuracy: 0.9513 - val_loss: 0.0990 - val_accuracy: 0.9713 Epoch 6/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1542 - accuracy: 0.9571 - val_loss: 0.0919 - val_accuracy: 0.9744 Epoch 7/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1487 - accuracy: 0.9599 - val_loss: 0.0844 - val_accuracy: 0.9766 Epoch 8/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1421 - accuracy: 0.9610 - val_loss: 0.0868 - val_accuracy: 0.9775 Epoch 9/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1322 - accuracy: 0.9642 - val_loss: 0.0804 - val_accuracy: 0.9793 Epoch 10/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1303 - accuracy: 0.9652 - val_loss: 0.0819 - val_accuracy: 0.9789 Epoch 11/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1248 - accuracy: 0.9668 - val_loss: 0.0845 - val_accuracy: 0.9786 Epoch 12/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1163 - accuracy: 0.9688 - val_loss: 0.0761 - val_accuracy: 0.9794 Epoch 13/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1180 - accuracy: 0.9686 - val_loss: 0.0745 - val_accuracy: 0.9821 Epoch 14/500 469/469 [==============================] - 2s 5ms/step - loss: 0.1127 - accuracy: 0.9694 - val_loss: 0.0754 - val_accuracy: 0.9816 Epoch 15/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1077 - accuracy: 0.9717 - val_loss: 0.0747 - val_accuracy: 0.9803 Epoch 16/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1090 - accuracy: 0.9707 - val_loss: 0.0712 - val_accuracy: 0.9826 Epoch 17/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1051 - accuracy: 0.9725 - val_loss: 0.0749 - val_accuracy: 0.9809 Epoch 18/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1020 - accuracy: 0.9728 - val_loss: 0.0716 - val_accuracy: 0.9802 Epoch 19/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0974 - accuracy: 0.9742 - val_loss: 0.0766 - val_accuracy: 0.9820 Epoch 20/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1045 - accuracy: 0.9732 - val_loss: 0.0761 - val_accuracy: 0.9819 Epoch 21/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1003 - accuracy: 0.9742 - val_loss: 0.0674 - val_accuracy: 0.9837 Epoch 22/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1001 - accuracy: 0.9746 - val_loss: 0.0738 - val_accuracy: 0.9826 Epoch 23/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0955 - accuracy: 0.9754 - val_loss: 0.0719 - val_accuracy: 0.9836 Epoch 24/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0926 - accuracy: 0.9757 - val_loss: 0.0661 - val_accuracy: 0.9834 Epoch 25/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0917 - accuracy: 0.9760 - val_loss: 0.0678 - val_accuracy: 0.9835 Epoch 26/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0873 - accuracy: 0.9772 - val_loss: 0.0761 - val_accuracy: 0.9839 Epoch 27/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0878 - accuracy: 0.9782 - val_loss: 0.0731 - val_accuracy: 0.9810 Epoch 28/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0887 - accuracy: 0.9771 - val_loss: 0.0763 - val_accuracy: 0.9824 Epoch 29/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0799 - accuracy: 0.9783 - val_loss: 0.0649 - val_accuracy: 0.9845 Epoch 30/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0785 - accuracy: 0.9790 - val_loss: 0.0733 - val_accuracy: 0.9825 Epoch 31/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0815 - accuracy: 0.9795 - val_loss: 0.0721 - val_accuracy: 0.9843 Epoch 32/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0827 - accuracy: 0.9785 - val_loss: 0.0653 - val_accuracy: 0.9836 Epoch 33/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0822 - accuracy: 0.9796 - val_loss: 0.0720 - val_accuracy: 0.9835 Epoch 34/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0804 - accuracy: 0.9797 - val_loss: 0.0621 - val_accuracy: 0.9852 Epoch 35/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0768 - accuracy: 0.9798 - val_loss: 0.0731 - val_accuracy: 0.9847 Epoch 36/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0834 - accuracy: 0.9797 - val_loss: 0.0693 - val_accuracy: 0.9858 Epoch 37/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0770 - accuracy: 0.9808 - val_loss: 0.0689 - val_accuracy: 0.9856 Epoch 38/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0756 - accuracy: 0.9812 - val_loss: 0.0731 - val_accuracy: 0.9838 Epoch 39/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0742 - accuracy: 0.9806 - val_loss: 0.0650 - val_accuracy: 0.9853 Epoch 40/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0763 - accuracy: 0.9812 - val_loss: 0.0778 - val_accuracy: 0.9838 Epoch 41/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0753 - accuracy: 0.9810 - val_loss: 0.0733 - val_accuracy: 0.9852 Epoch 42/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0706 - accuracy: 0.9819 - val_loss: 0.0738 - val_accuracy: 0.9843 Epoch 43/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0705 - accuracy: 0.9824 - val_loss: 0.0712 - val_accuracy: 0.9842 Epoch 44/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0724 - accuracy: 0.9829 - val_loss: 0.0702 - val_accuracy: 0.9848 Epoch 45/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0694 - accuracy: 0.9822 - val_loss: 0.0733 - val_accuracy: 0.9854 Epoch 46/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0724 - accuracy: 0.9821 - val_loss: 0.0718 - val_accuracy: 0.9846 Epoch 47/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0782 - accuracy: 0.9819 - val_loss: 0.0729 - val_accuracy: 0.9857 Epoch 48/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0731 - accuracy: 0.9821 - val_loss: 0.0732 - val_accuracy: 0.9848 Epoch 49/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0671 - accuracy: 0.9840 - val_loss: 0.0701 - val_accuracy: 0.9856 Epoch 50/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0656 - accuracy: 0.9833 - val_loss: 0.0642 - val_accuracy: 0.9864 Epoch 51/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0700 - accuracy: 0.9832 - val_loss: 0.0725 - val_accuracy: 0.9845 Epoch 52/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0646 - accuracy: 0.9841 - val_loss: 0.0756 - val_accuracy: 0.9848 Epoch 53/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0688 - accuracy: 0.9837 - val_loss: 0.0789 - val_accuracy: 0.9865 Epoch 54/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0674 - accuracy: 0.9841 - val_loss: 0.0671 - val_accuracy: 0.9859 Epoch 55/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0659 - accuracy: 0.9835 - val_loss: 0.0669 - val_accuracy: 0.9870 Epoch 56/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0643 - accuracy: 0.9840 - val_loss: 0.0796 - val_accuracy: 0.9857 Epoch 57/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0655 - accuracy: 0.9842 - val_loss: 0.0704 - val_accuracy: 0.9851 Epoch 58/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0629 - accuracy: 0.9848 - val_loss: 0.0648 - val_accuracy: 0.9871 Epoch 59/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0617 - accuracy: 0.9846 - val_loss: 0.0645 - val_accuracy: 0.9864 Epoch 60/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0586 - accuracy: 0.9856 - val_loss: 0.0707 - val_accuracy: 0.9870 Epoch 61/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0691 - accuracy: 0.9838 - val_loss: 0.0706 - val_accuracy: 0.9865 Epoch 62/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0644 - accuracy: 0.9845 - val_loss: 0.0799 - val_accuracy: 0.9852 Epoch 63/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0602 - accuracy: 0.9854 - val_loss: 0.0774 - val_accuracy: 0.9849 Epoch 64/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0643 - accuracy: 0.9845 - val_loss: 0.0799 - val_accuracy: 0.9846 Epoch 65/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0624 - accuracy: 0.9859 - val_loss: 0.0659 - val_accuracy: 0.9862 Epoch 66/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0677 - accuracy: 0.9845 - val_loss: 0.0640 - val_accuracy: 0.9861 Epoch 67/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0679 - accuracy: 0.9851 - val_loss: 0.0773 - val_accuracy: 0.9865 Epoch 68/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0644 - accuracy: 0.9855 - val_loss: 0.0657 - val_accuracy: 0.9869 Epoch 69/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0635 - accuracy: 0.9848 - val_loss: 0.0703 - val_accuracy: 0.9854 Epoch 70/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0600 - accuracy: 0.9856 - val_loss: 0.0670 - val_accuracy: 0.9853 Epoch 71/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0574 - accuracy: 0.9859 - val_loss: 0.0778 - val_accuracy: 0.9856 Epoch 72/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0564 - accuracy: 0.9868 - val_loss: 0.0673 - val_accuracy: 0.9868 Epoch 73/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0575 - accuracy: 0.9866 - val_loss: 0.0688 - val_accuracy: 0.9868 Epoch 74/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0598 - accuracy: 0.9866 - val_loss: 0.0613 - val_accuracy: 0.9874 Epoch 75/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0638 - accuracy: 0.9853 - val_loss: 0.0670 - val_accuracy: 0.9868 Epoch 76/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0635 - accuracy: 0.9857 - val_loss: 0.0583 - val_accuracy: 0.9865 Epoch 77/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0630 - accuracy: 0.9856 - val_loss: 0.0649 - val_accuracy: 0.9855 Epoch 78/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0589 - accuracy: 0.9865 - val_loss: 0.0676 - val_accuracy: 0.9852 Epoch 79/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0580 - accuracy: 0.9868 - val_loss: 0.0763 - val_accuracy: 0.9859 Epoch 80/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0645 - accuracy: 0.9859 - val_loss: 0.0714 - val_accuracy: 0.9864 Epoch 81/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0520 - accuracy: 0.9877 - val_loss: 0.0664 - val_accuracy: 0.9857 Epoch 82/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0583 - accuracy: 0.9870 - val_loss: 0.0644 - val_accuracy: 0.9870 Epoch 83/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0627 - accuracy: 0.9865 - val_loss: 0.0766 - val_accuracy: 0.9861 Epoch 84/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0613 - accuracy: 0.9859 - val_loss: 0.0760 - val_accuracy: 0.9862 Epoch 85/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0619 - accuracy: 0.9866 - val_loss: 0.0683 - val_accuracy: 0.9856 Epoch 86/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0545 - accuracy: 0.9874 - val_loss: 0.0717 - val_accuracy: 0.9866 Epoch 87/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0560 - accuracy: 0.9868 - val_loss: 0.0791 - val_accuracy: 0.9871 Epoch 88/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0514 - accuracy: 0.9880 - val_loss: 0.0749 - val_accuracy: 0.9851 Epoch 89/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9879 - val_loss: 0.0719 - val_accuracy: 0.9851 Epoch 90/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0620 - accuracy: 0.9862 - val_loss: 0.0694 - val_accuracy: 0.9856 Epoch 91/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0597 - accuracy: 0.9869 - val_loss: 0.0673 - val_accuracy: 0.9855 Epoch 92/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0532 - accuracy: 0.9878 - val_loss: 0.0692 - val_accuracy: 0.9862 Epoch 93/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0550 - accuracy: 0.9872 - val_loss: 0.0765 - val_accuracy: 0.9854 Epoch 94/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0547 - accuracy: 0.9876 - val_loss: 0.0647 - val_accuracy: 0.9871 Epoch 95/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0568 - accuracy: 0.9878 - val_loss: 0.0677 - val_accuracy: 0.9846 Epoch 96/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0546 - accuracy: 0.9875 - val_loss: 0.0777 - val_accuracy: 0.9851 Epoch 97/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0542 - accuracy: 0.9879 - val_loss: 0.0729 - val_accuracy: 0.9835 Epoch 98/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0620 - accuracy: 0.9872 - val_loss: 0.0670 - val_accuracy: 0.9853 Epoch 99/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0575 - accuracy: 0.9869 - val_loss: 0.0674 - val_accuracy: 0.9864 Epoch 100/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0597 - accuracy: 0.9876 - val_loss: 0.0680 - val_accuracy: 0.9847 Epoch 101/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0517 - accuracy: 0.9887 - val_loss: 0.0693 - val_accuracy: 0.9874 Epoch 102/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0534 - accuracy: 0.9878 - val_loss: 0.0717 - val_accuracy: 0.9858 Epoch 103/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0564 - accuracy: 0.9871 - val_loss: 0.0693 - val_accuracy: 0.9862 Epoch 104/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0519 - accuracy: 0.9882 - val_loss: 0.0658 - val_accuracy: 0.9851 Epoch 105/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0545 - accuracy: 0.9885 - val_loss: 0.0694 - val_accuracy: 0.9868 Epoch 106/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0617 - accuracy: 0.9878 - val_loss: 0.0693 - val_accuracy: 0.9847 Epoch 107/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0563 - accuracy: 0.9877 - val_loss: 0.0669 - val_accuracy: 0.9856 Epoch 108/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0567 - accuracy: 0.9881 - val_loss: 0.0772 - val_accuracy: 0.9849 Epoch 109/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0504 - accuracy: 0.9886 - val_loss: 0.0681 - val_accuracy: 0.9868 Epoch 110/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0573 - accuracy: 0.9877 - val_loss: 0.0693 - val_accuracy: 0.9864 Epoch 111/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0503 - accuracy: 0.9885 - val_loss: 0.0859 - val_accuracy: 0.9854 Epoch 112/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0515 - accuracy: 0.9887 - val_loss: 0.0656 - val_accuracy: 0.9868 Epoch 113/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0583 - accuracy: 0.9878 - val_loss: 0.0687 - val_accuracy: 0.9865 Epoch 114/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0564 - accuracy: 0.9879 - val_loss: 0.0622 - val_accuracy: 0.9876 Epoch 115/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0564 - accuracy: 0.9876 - val_loss: 0.0779 - val_accuracy: 0.9869 Epoch 116/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0517 - accuracy: 0.9887 - val_loss: 0.0701 - val_accuracy: 0.9870 Epoch 117/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0534 - accuracy: 0.9891 - val_loss: 0.0767 - val_accuracy: 0.9864 Epoch 118/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0561 - accuracy: 0.9881 - val_loss: 0.0652 - val_accuracy: 0.9866 Epoch 119/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0510 - accuracy: 0.9883 - val_loss: 0.0668 - val_accuracy: 0.9864 Epoch 120/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0546 - accuracy: 0.9882 - val_loss: 0.0646 - val_accuracy: 0.9873 Epoch 121/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0566 - accuracy: 0.9883 - val_loss: 0.0740 - val_accuracy: 0.9855 Epoch 122/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0525 - accuracy: 0.9891 - val_loss: 0.0723 - val_accuracy: 0.9873 Epoch 123/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0567 - accuracy: 0.9883 - val_loss: 0.0672 - val_accuracy: 0.9865 Epoch 124/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0571 - accuracy: 0.9893 - val_loss: 0.0696 - val_accuracy: 0.9855 Epoch 125/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0554 - accuracy: 0.9889 - val_loss: 0.0752 - val_accuracy: 0.9852 Epoch 126/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0642 - accuracy: 0.9884 - val_loss: 0.0738 - val_accuracy: 0.9837 Epoch 127/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0541 - accuracy: 0.9890 - val_loss: 0.0704 - val_accuracy: 0.9862 Epoch 128/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0532 - accuracy: 0.9892 - val_loss: 0.0813 - val_accuracy: 0.9853 Epoch 129/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0609 - accuracy: 0.9885 - val_loss: 0.0739 - val_accuracy: 0.9862 Epoch 130/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0548 - accuracy: 0.9889 - val_loss: 0.0686 - val_accuracy: 0.9872 Epoch 131/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0597 - accuracy: 0.9895 - val_loss: 0.0757 - val_accuracy: 0.9859 Epoch 132/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0507 - accuracy: 0.9885 - val_loss: 0.0707 - val_accuracy: 0.9852 Epoch 133/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9890 - val_loss: 0.0659 - val_accuracy: 0.9864 Epoch 134/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0527 - accuracy: 0.9894 - val_loss: 0.0711 - val_accuracy: 0.9850 Epoch 135/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9892 - val_loss: 0.0721 - val_accuracy: 0.9854 Epoch 136/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0510 - accuracy: 0.9897 - val_loss: 0.0713 - val_accuracy: 0.9870 Epoch 137/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0519 - accuracy: 0.9890 - val_loss: 0.0815 - val_accuracy: 0.9869 Epoch 138/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0553 - accuracy: 0.9883 - val_loss: 0.0826 - val_accuracy: 0.9859 Epoch 139/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0529 - accuracy: 0.9892 - val_loss: 0.0734 - val_accuracy: 0.9846 Epoch 140/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0584 - accuracy: 0.9886 - val_loss: 0.0821 - val_accuracy: 0.9860 Epoch 141/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0553 - accuracy: 0.9886 - val_loss: 0.0722 - val_accuracy: 0.9869 Epoch 142/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0647 - accuracy: 0.9882 - val_loss: 0.0758 - val_accuracy: 0.9859 Epoch 143/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0597 - accuracy: 0.9886 - val_loss: 0.0726 - val_accuracy: 0.9860 Epoch 144/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0513 - accuracy: 0.9895 - val_loss: 0.0734 - val_accuracy: 0.9865 Epoch 145/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0576 - accuracy: 0.9891 - val_loss: 0.0769 - val_accuracy: 0.9857 Epoch 146/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0640 - accuracy: 0.9880 - val_loss: 0.0752 - val_accuracy: 0.9861 Epoch 147/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0510 - accuracy: 0.9899 - val_loss: 0.0699 - val_accuracy: 0.9871 Epoch 148/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0556 - accuracy: 0.9894 - val_loss: 0.0709 - val_accuracy: 0.9852 Epoch 149/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0555 - accuracy: 0.9893 - val_loss: 0.0732 - val_accuracy: 0.9866 Epoch 150/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0553 - accuracy: 0.9895 - val_loss: 0.0849 - val_accuracy: 0.9860 Epoch 151/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0520 - accuracy: 0.9899 - val_loss: 0.0760 - val_accuracy: 0.9868 Epoch 152/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0621 - accuracy: 0.9893 - val_loss: 0.0859 - val_accuracy: 0.9863 Epoch 153/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0622 - accuracy: 0.9882 - val_loss: 0.0726 - val_accuracy: 0.9854 Epoch 154/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0545 - accuracy: 0.9894 - val_loss: 0.0829 - val_accuracy: 0.9854 Epoch 155/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0575 - accuracy: 0.9890 - val_loss: 0.0774 - val_accuracy: 0.9854 Epoch 156/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0499 - accuracy: 0.9898 - val_loss: 0.0914 - val_accuracy: 0.9864 Epoch 157/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0512 - accuracy: 0.9892 - val_loss: 0.0742 - val_accuracy: 0.9870 Epoch 158/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0543 - accuracy: 0.9897 - val_loss: 0.0753 - val_accuracy: 0.9875 Epoch 159/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0469 - accuracy: 0.9910 - val_loss: 0.0715 - val_accuracy: 0.9864 Epoch 160/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0496 - accuracy: 0.9899 - val_loss: 0.0813 - val_accuracy: 0.9874 Epoch 161/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0564 - accuracy: 0.9894 - val_loss: 0.0748 - val_accuracy: 0.9859 Epoch 162/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0499 - accuracy: 0.9895 - val_loss: 0.0772 - val_accuracy: 0.9865 Epoch 163/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0465 - accuracy: 0.9902 - val_loss: 0.0772 - val_accuracy: 0.9861 Epoch 164/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0613 - accuracy: 0.9891 - val_loss: 0.0971 - val_accuracy: 0.9858 Epoch 165/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0665 - accuracy: 0.9891 - val_loss: 0.0859 - val_accuracy: 0.9860 Epoch 166/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0545 - accuracy: 0.9897 - val_loss: 0.0858 - val_accuracy: 0.9854 Epoch 167/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0571 - accuracy: 0.9894 - val_loss: 0.0821 - val_accuracy: 0.9870 Epoch 168/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0545 - accuracy: 0.9906 - val_loss: 0.0753 - val_accuracy: 0.9873 Epoch 169/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0598 - accuracy: 0.9896 - val_loss: 0.0803 - val_accuracy: 0.9859 Epoch 170/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0547 - accuracy: 0.9896 - val_loss: 0.0699 - val_accuracy: 0.9852 Epoch 171/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0484 - accuracy: 0.9901 - val_loss: 0.0729 - val_accuracy: 0.9854 Epoch 172/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0472 - accuracy: 0.9905 - val_loss: 0.0777 - val_accuracy: 0.9866 Epoch 173/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0536 - accuracy: 0.9901 - val_loss: 0.0778 - val_accuracy: 0.9845 Epoch 174/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0545 - accuracy: 0.9896 - val_loss: 0.0942 - val_accuracy: 0.9855 Epoch 175/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0620 - accuracy: 0.9897 - val_loss: 0.0912 - val_accuracy: 0.9855 Epoch 176/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0612 - accuracy: 0.9888 - val_loss: 0.0814 - val_accuracy: 0.9865 Epoch 177/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0511 - accuracy: 0.9900 - val_loss: 0.0795 - val_accuracy: 0.9861 Epoch 178/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0558 - accuracy: 0.9895 - val_loss: 0.0778 - val_accuracy: 0.9864 Epoch 179/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0520 - accuracy: 0.9894 - val_loss: 0.0689 - val_accuracy: 0.9887 Epoch 180/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0463 - accuracy: 0.9908 - val_loss: 0.0771 - val_accuracy: 0.9863 Epoch 181/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0606 - accuracy: 0.9891 - val_loss: 0.0810 - val_accuracy: 0.9869 Epoch 182/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0480 - accuracy: 0.9903 - val_loss: 0.1135 - val_accuracy: 0.9854 Epoch 183/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0619 - accuracy: 0.9891 - val_loss: 0.0744 - val_accuracy: 0.9848 Epoch 184/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0514 - accuracy: 0.9901 - val_loss: 0.0749 - val_accuracy: 0.9867 Epoch 185/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0480 - accuracy: 0.9901 - val_loss: 0.0870 - val_accuracy: 0.9864 Epoch 186/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0485 - accuracy: 0.9910 - val_loss: 0.0902 - val_accuracy: 0.9865 Epoch 187/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0467 - accuracy: 0.9907 - val_loss: 0.0758 - val_accuracy: 0.9860 Epoch 188/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0500 - accuracy: 0.9901 - val_loss: 0.0875 - val_accuracy: 0.9874 Epoch 189/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0551 - accuracy: 0.9904 - val_loss: 0.0714 - val_accuracy: 0.9876 Epoch 190/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0596 - accuracy: 0.9904 - val_loss: 0.0843 - val_accuracy: 0.9860 Epoch 191/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0613 - accuracy: 0.9896 - val_loss: 0.0836 - val_accuracy: 0.9862 Epoch 192/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0554 - accuracy: 0.9898 - val_loss: 0.0898 - val_accuracy: 0.9866 Epoch 193/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0516 - accuracy: 0.9908 - val_loss: 0.0896 - val_accuracy: 0.9864 Epoch 194/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0576 - accuracy: 0.9901 - val_loss: 0.0837 - val_accuracy: 0.9869 Epoch 195/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0552 - accuracy: 0.9905 - val_loss: 0.0785 - val_accuracy: 0.9870 Epoch 196/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0509 - accuracy: 0.9905 - val_loss: 0.0881 - val_accuracy: 0.9850 Epoch 197/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0499 - accuracy: 0.9907 - val_loss: 0.0891 - val_accuracy: 0.9866 Epoch 198/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0453 - accuracy: 0.9908 - val_loss: 0.0854 - val_accuracy: 0.9867 Epoch 199/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0490 - accuracy: 0.9910 - val_loss: 0.0809 - val_accuracy: 0.9859 Epoch 200/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0500 - accuracy: 0.9907 - val_loss: 0.0912 - val_accuracy: 0.9856 Epoch 201/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0494 - accuracy: 0.9903 - val_loss: 0.0871 - val_accuracy: 0.9864 Epoch 202/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0551 - accuracy: 0.9904 - val_loss: 0.0737 - val_accuracy: 0.9867 Epoch 203/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9904 - val_loss: 0.0837 - val_accuracy: 0.9867 Epoch 204/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0510 - accuracy: 0.9906 - val_loss: 0.0822 - val_accuracy: 0.9868 Epoch 205/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0491 - accuracy: 0.9910 - val_loss: 0.0821 - val_accuracy: 0.9868 Epoch 206/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0466 - accuracy: 0.9911 - val_loss: 0.0737 - val_accuracy: 0.9869 Epoch 207/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0486 - accuracy: 0.9906 - val_loss: 0.0806 - val_accuracy: 0.9864 Epoch 208/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0569 - accuracy: 0.9907 - val_loss: 0.0887 - val_accuracy: 0.9852 Epoch 209/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0521 - accuracy: 0.9910 - val_loss: 0.0753 - val_accuracy: 0.9870 Epoch 210/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0528 - accuracy: 0.9905 - val_loss: 0.0911 - val_accuracy: 0.9850 Epoch 211/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0588 - accuracy: 0.9903 - val_loss: 0.0844 - val_accuracy: 0.9858 Epoch 212/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0608 - accuracy: 0.9905 - val_loss: 0.0898 - val_accuracy: 0.9866 Epoch 213/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0409 - accuracy: 0.9917 - val_loss: 0.0772 - val_accuracy: 0.9865 Epoch 214/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0560 - accuracy: 0.9898 - val_loss: 0.0724 - val_accuracy: 0.9877 Epoch 215/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0506 - accuracy: 0.9908 - val_loss: 0.0741 - val_accuracy: 0.9862 Epoch 216/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0468 - accuracy: 0.9912 - val_loss: 0.0937 - val_accuracy: 0.9858 Epoch 217/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0545 - accuracy: 0.9900 - val_loss: 0.0875 - val_accuracy: 0.9856 Epoch 218/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0516 - accuracy: 0.9907 - val_loss: 0.0872 - val_accuracy: 0.9858 Epoch 219/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0564 - accuracy: 0.9904 - val_loss: 0.0829 - val_accuracy: 0.9863 Epoch 220/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0531 - accuracy: 0.9905 - val_loss: 0.0910 - val_accuracy: 0.9859 Epoch 221/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0492 - accuracy: 0.9912 - val_loss: 0.0818 - val_accuracy: 0.9864 Epoch 222/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0498 - accuracy: 0.9908 - val_loss: 0.0837 - val_accuracy: 0.9860 Epoch 223/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0513 - accuracy: 0.9908 - val_loss: 0.0755 - val_accuracy: 0.9869 Epoch 224/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0519 - accuracy: 0.9910 - val_loss: 0.0771 - val_accuracy: 0.9862 Epoch 225/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0473 - accuracy: 0.9919 - val_loss: 0.0787 - val_accuracy: 0.9853 Epoch 226/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0437 - accuracy: 0.9923 - val_loss: 0.0839 - val_accuracy: 0.9870 Epoch 227/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0581 - accuracy: 0.9907 - val_loss: 0.0802 - val_accuracy: 0.9865 Epoch 228/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0545 - accuracy: 0.9905 - val_loss: 0.0794 - val_accuracy: 0.9865 Epoch 229/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0485 - accuracy: 0.9909 - val_loss: 0.0866 - val_accuracy: 0.9864 Epoch 230/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0586 - accuracy: 0.9909 - val_loss: 0.0920 - val_accuracy: 0.9851 Epoch 231/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0572 - accuracy: 0.9905 - val_loss: 0.0876 - val_accuracy: 0.9853 Epoch 232/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0577 - accuracy: 0.9908 - val_loss: 0.0776 - val_accuracy: 0.9861 Epoch 233/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0499 - accuracy: 0.9907 - val_loss: 0.0833 - val_accuracy: 0.9867 Epoch 234/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0471 - accuracy: 0.9919 - val_loss: 0.0866 - val_accuracy: 0.9864 Epoch 235/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0591 - accuracy: 0.9903 - val_loss: 0.0763 - val_accuracy: 0.9869 Epoch 236/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0505 - accuracy: 0.9903 - val_loss: 0.0846 - val_accuracy: 0.9880 Epoch 237/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0585 - accuracy: 0.9902 - val_loss: 0.0726 - val_accuracy: 0.9870 Epoch 238/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0617 - accuracy: 0.9903 - val_loss: 0.0748 - val_accuracy: 0.9872 Epoch 239/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0536 - accuracy: 0.9909 - val_loss: 0.0727 - val_accuracy: 0.9875 Epoch 240/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0595 - accuracy: 0.9905 - val_loss: 0.0717 - val_accuracy: 0.9869 Epoch 241/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0513 - accuracy: 0.9914 - val_loss: 0.0802 - val_accuracy: 0.9867 Epoch 242/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0545 - accuracy: 0.9904 - val_loss: 0.0857 - val_accuracy: 0.9871 Epoch 243/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0525 - accuracy: 0.9908 - val_loss: 0.0699 - val_accuracy: 0.9869 Epoch 244/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0487 - accuracy: 0.9914 - val_loss: 0.0898 - val_accuracy: 0.9869 Epoch 245/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0504 - accuracy: 0.9915 - val_loss: 0.0762 - val_accuracy: 0.9869 Epoch 246/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0479 - accuracy: 0.9914 - val_loss: 0.0809 - val_accuracy: 0.9878 Epoch 247/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0500 - accuracy: 0.9908 - val_loss: 0.0842 - val_accuracy: 0.9869 Epoch 248/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0467 - accuracy: 0.9918 - val_loss: 0.0795 - val_accuracy: 0.9876 Epoch 249/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0779 - accuracy: 0.9904 - val_loss: 0.0895 - val_accuracy: 0.9869 Epoch 250/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0595 - accuracy: 0.9905 - val_loss: 0.0927 - val_accuracy: 0.9872 Epoch 251/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0544 - accuracy: 0.9907 - val_loss: 0.0774 - val_accuracy: 0.9882 Epoch 252/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0483 - accuracy: 0.9915 - val_loss: 0.0740 - val_accuracy: 0.9873 Epoch 253/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0538 - accuracy: 0.9903 - val_loss: 0.0865 - val_accuracy: 0.9865 Epoch 254/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0539 - accuracy: 0.9906 - val_loss: 0.0967 - val_accuracy: 0.9856 Epoch 255/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0589 - accuracy: 0.9904 - val_loss: 0.0899 - val_accuracy: 0.9878 Epoch 256/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0461 - accuracy: 0.9910 - val_loss: 0.0733 - val_accuracy: 0.9867 Epoch 257/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0565 - accuracy: 0.9911 - val_loss: 0.0869 - val_accuracy: 0.9871 Epoch 258/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0514 - accuracy: 0.9909 - val_loss: 0.0819 - val_accuracy: 0.9878 Epoch 259/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0575 - accuracy: 0.9916 - val_loss: 0.0735 - val_accuracy: 0.9871 Epoch 260/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0487 - accuracy: 0.9914 - val_loss: 0.0938 - val_accuracy: 0.9872 Epoch 261/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0631 - accuracy: 0.9892 - val_loss: 0.0939 - val_accuracy: 0.9860 Epoch 262/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9905 - val_loss: 0.0839 - val_accuracy: 0.9880 Epoch 263/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0462 - accuracy: 0.9913 - val_loss: 0.0788 - val_accuracy: 0.9874 Epoch 264/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0563 - accuracy: 0.9910 - val_loss: 0.0938 - val_accuracy: 0.9874 Epoch 265/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0487 - accuracy: 0.9912 - val_loss: 0.0755 - val_accuracy: 0.9883 Epoch 266/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0611 - accuracy: 0.9901 - val_loss: 0.0796 - val_accuracy: 0.9874 Epoch 267/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0458 - accuracy: 0.9918 - val_loss: 0.0750 - val_accuracy: 0.9877 Epoch 268/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0525 - accuracy: 0.9913 - val_loss: 0.0780 - val_accuracy: 0.9876 Epoch 269/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0494 - accuracy: 0.9909 - val_loss: 0.0847 - val_accuracy: 0.9866 Epoch 270/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0496 - accuracy: 0.9909 - val_loss: 0.0965 - val_accuracy: 0.9855 Epoch 271/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0502 - accuracy: 0.9917 - val_loss: 0.0939 - val_accuracy: 0.9861 Epoch 272/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0504 - accuracy: 0.9913 - val_loss: 0.0918 - val_accuracy: 0.9872 Epoch 273/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0462 - accuracy: 0.9917 - val_loss: 0.0929 - val_accuracy: 0.9880 Epoch 274/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0552 - accuracy: 0.9914 - val_loss: 0.0749 - val_accuracy: 0.9866 Epoch 275/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0525 - accuracy: 0.9918 - val_loss: 0.0894 - val_accuracy: 0.9866 Epoch 276/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0665 - accuracy: 0.9905 - val_loss: 0.0892 - val_accuracy: 0.9869 Epoch 277/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0586 - accuracy: 0.9907 - val_loss: 0.0762 - val_accuracy: 0.9872 Epoch 278/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0495 - accuracy: 0.9911 - val_loss: 0.0822 - val_accuracy: 0.9866 Epoch 279/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0437 - accuracy: 0.9917 - val_loss: 0.0781 - val_accuracy: 0.9862 Epoch 280/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0483 - accuracy: 0.9914 - val_loss: 0.0823 - val_accuracy: 0.9867 Epoch 281/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0592 - accuracy: 0.9922 - val_loss: 0.0918 - val_accuracy: 0.9876 Epoch 282/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0675 - accuracy: 0.9912 - val_loss: 0.0937 - val_accuracy: 0.9875 Epoch 283/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0520 - accuracy: 0.9913 - val_loss: 0.0816 - val_accuracy: 0.9878 Epoch 284/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0523 - accuracy: 0.9919 - val_loss: 0.0866 - val_accuracy: 0.9869 Epoch 285/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0470 - accuracy: 0.9913 - val_loss: 0.0841 - val_accuracy: 0.9865 Epoch 286/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0519 - accuracy: 0.9916 - val_loss: 0.0994 - val_accuracy: 0.9869 Epoch 287/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0463 - accuracy: 0.9915 - val_loss: 0.0868 - val_accuracy: 0.9870 Epoch 288/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0518 - accuracy: 0.9913 - val_loss: 0.0769 - val_accuracy: 0.9867 Epoch 289/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0487 - accuracy: 0.9918 - val_loss: 0.0756 - val_accuracy: 0.9874 Epoch 290/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0772 - accuracy: 0.9907 - val_loss: 0.0807 - val_accuracy: 0.9861 Epoch 291/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0513 - accuracy: 0.9909 - val_loss: 0.1009 - val_accuracy: 0.9861 Epoch 292/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0599 - accuracy: 0.9907 - val_loss: 0.0779 - val_accuracy: 0.9868 Epoch 293/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0535 - accuracy: 0.9912 - val_loss: 0.0880 - val_accuracy: 0.9867 Epoch 294/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0614 - accuracy: 0.9903 - val_loss: 0.0870 - val_accuracy: 0.9864 Epoch 295/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0539 - accuracy: 0.9910 - val_loss: 0.0885 - val_accuracy: 0.9859 Epoch 296/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0467 - accuracy: 0.9915 - val_loss: 0.0795 - val_accuracy: 0.9864 Epoch 297/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0617 - accuracy: 0.9912 - val_loss: 0.0851 - val_accuracy: 0.9861 Epoch 298/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0632 - accuracy: 0.9907 - val_loss: 0.0880 - val_accuracy: 0.9863 Epoch 299/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0520 - accuracy: 0.9912 - val_loss: 0.1030 - val_accuracy: 0.9864 Epoch 300/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0431 - accuracy: 0.9925 - val_loss: 0.0932 - val_accuracy: 0.9871 Epoch 301/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0504 - accuracy: 0.9913 - val_loss: 0.0898 - val_accuracy: 0.9883 Epoch 302/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0427 - accuracy: 0.9923 - val_loss: 0.1005 - val_accuracy: 0.9876 Epoch 303/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0402 - accuracy: 0.9923 - val_loss: 0.0861 - val_accuracy: 0.9881 Epoch 304/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0488 - accuracy: 0.9919 - val_loss: 0.0791 - val_accuracy: 0.9868 Epoch 305/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0501 - accuracy: 0.9919 - val_loss: 0.0993 - val_accuracy: 0.9872 Epoch 306/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0553 - accuracy: 0.9918 - val_loss: 0.0954 - val_accuracy: 0.9872 Epoch 307/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0779 - accuracy: 0.9911 - val_loss: 0.0840 - val_accuracy: 0.9861 Epoch 308/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0568 - accuracy: 0.9902 - val_loss: 0.0894 - val_accuracy: 0.9883 Epoch 309/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0515 - accuracy: 0.9908 - val_loss: 0.0912 - val_accuracy: 0.9866 Epoch 310/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0424 - accuracy: 0.9921 - val_loss: 0.0938 - val_accuracy: 0.9875 Epoch 311/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0554 - accuracy: 0.9919 - val_loss: 0.0887 - val_accuracy: 0.9881 Epoch 312/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0447 - accuracy: 0.9923 - val_loss: 0.1094 - val_accuracy: 0.9867 Epoch 313/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0544 - accuracy: 0.9919 - val_loss: 0.0923 - val_accuracy: 0.9879 Epoch 314/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0585 - accuracy: 0.9911 - val_loss: 0.0886 - val_accuracy: 0.9880 Epoch 315/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0816 - accuracy: 0.9911 - val_loss: 0.0923 - val_accuracy: 0.9870 Epoch 316/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0663 - accuracy: 0.9905 - val_loss: 0.0960 - val_accuracy: 0.9870 Epoch 317/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0467 - accuracy: 0.9912 - val_loss: 0.0895 - val_accuracy: 0.9875 Epoch 318/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0454 - accuracy: 0.9920 - val_loss: 0.0989 - val_accuracy: 0.9877 Epoch 319/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0507 - accuracy: 0.9919 - val_loss: 0.0911 - val_accuracy: 0.9875 Epoch 320/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0491 - accuracy: 0.9918 - val_loss: 0.0996 - val_accuracy: 0.9881 Epoch 321/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0435 - accuracy: 0.9921 - val_loss: 0.1045 - val_accuracy: 0.9873 Epoch 322/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0500 - accuracy: 0.9913 - val_loss: 0.1012 - val_accuracy: 0.9868 Epoch 323/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0542 - accuracy: 0.9920 - val_loss: 0.1033 - val_accuracy: 0.9862 Epoch 324/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0578 - accuracy: 0.9918 - val_loss: 0.0973 - val_accuracy: 0.9869 Epoch 325/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1249 - accuracy: 0.9902 - val_loss: 0.0819 - val_accuracy: 0.9880 Epoch 326/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0490 - accuracy: 0.9904 - val_loss: 0.0904 - val_accuracy: 0.9872 Epoch 327/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0523 - accuracy: 0.9913 - val_loss: 0.0856 - val_accuracy: 0.9879 Epoch 328/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0626 - accuracy: 0.9917 - val_loss: 0.0915 - val_accuracy: 0.9878 Epoch 329/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0805 - accuracy: 0.9913 - val_loss: 0.0997 - val_accuracy: 0.9875 Epoch 330/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0527 - accuracy: 0.9903 - val_loss: 0.0841 - val_accuracy: 0.9881 Epoch 331/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0506 - accuracy: 0.9920 - val_loss: 0.0915 - val_accuracy: 0.9886 Epoch 332/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0914 - accuracy: 0.9920 - val_loss: 0.0953 - val_accuracy: 0.9883 Epoch 333/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0538 - accuracy: 0.9915 - val_loss: 0.1008 - val_accuracy: 0.9883 Epoch 334/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0565 - accuracy: 0.9916 - val_loss: 0.0867 - val_accuracy: 0.9885 Epoch 335/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0509 - accuracy: 0.9912 - val_loss: 0.0876 - val_accuracy: 0.9876 Epoch 336/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0509 - accuracy: 0.9917 - val_loss: 0.0820 - val_accuracy: 0.9880 Epoch 337/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0416 - accuracy: 0.9926 - val_loss: 0.1034 - val_accuracy: 0.9882 Epoch 338/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0482 - accuracy: 0.9922 - val_loss: 0.0744 - val_accuracy: 0.9877 Epoch 339/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0542 - accuracy: 0.9923 - val_loss: 0.0893 - val_accuracy: 0.9869 Epoch 340/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0497 - accuracy: 0.9922 - val_loss: 0.0929 - val_accuracy: 0.9874 Epoch 341/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0471 - accuracy: 0.9922 - val_loss: 0.0854 - val_accuracy: 0.9881 Epoch 342/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0564 - accuracy: 0.9916 - val_loss: 0.0984 - val_accuracy: 0.9884 Epoch 343/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0624 - accuracy: 0.9906 - val_loss: 0.1047 - val_accuracy: 0.9882 Epoch 344/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0429 - accuracy: 0.9920 - val_loss: 0.0863 - val_accuracy: 0.9877 Epoch 345/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0597 - accuracy: 0.9922 - val_loss: 0.0736 - val_accuracy: 0.9885 Epoch 346/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0567 - accuracy: 0.9917 - val_loss: 0.0880 - val_accuracy: 0.9875 Epoch 347/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0491 - accuracy: 0.9913 - val_loss: 0.0826 - val_accuracy: 0.9869 Epoch 348/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0458 - accuracy: 0.9916 - val_loss: 0.1060 - val_accuracy: 0.9871 Epoch 349/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0440 - accuracy: 0.9924 - val_loss: 0.1251 - val_accuracy: 0.9874 Epoch 350/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0565 - accuracy: 0.9924 - val_loss: 0.0974 - val_accuracy: 0.9878 Epoch 351/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0521 - accuracy: 0.9917 - val_loss: 0.1117 - val_accuracy: 0.9866 Epoch 352/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0615 - accuracy: 0.9911 - val_loss: 0.1331 - val_accuracy: 0.9871 Epoch 353/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0536 - accuracy: 0.9910 - val_loss: 0.1073 - val_accuracy: 0.9869 Epoch 354/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0481 - accuracy: 0.9924 - val_loss: 0.0826 - val_accuracy: 0.9876 Epoch 355/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0418 - accuracy: 0.9923 - val_loss: 0.0935 - val_accuracy: 0.9869 Epoch 356/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0481 - accuracy: 0.9920 - val_loss: 0.1087 - val_accuracy: 0.9879 Epoch 357/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0614 - accuracy: 0.9915 - val_loss: 0.0867 - val_accuracy: 0.9872 Epoch 358/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0682 - accuracy: 0.9903 - val_loss: 0.1048 - val_accuracy: 0.9869 Epoch 359/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0508 - accuracy: 0.9918 - val_loss: 0.1055 - val_accuracy: 0.9872 Epoch 360/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0437 - accuracy: 0.9920 - val_loss: 0.0962 - val_accuracy: 0.9868 Epoch 361/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0657 - accuracy: 0.9918 - val_loss: 0.0961 - val_accuracy: 0.9866 Epoch 362/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0748 - accuracy: 0.9902 - val_loss: 0.1036 - val_accuracy: 0.9883 Epoch 363/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0746 - accuracy: 0.9902 - val_loss: 0.0903 - val_accuracy: 0.9878 Epoch 364/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0477 - accuracy: 0.9916 - val_loss: 0.0794 - val_accuracy: 0.9888 Epoch 365/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0458 - accuracy: 0.9925 - val_loss: 0.0788 - val_accuracy: 0.9875 Epoch 366/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0469 - accuracy: 0.9927 - val_loss: 0.0870 - val_accuracy: 0.9883 Epoch 367/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0543 - accuracy: 0.9916 - val_loss: 0.0983 - val_accuracy: 0.9864 Epoch 368/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0440 - accuracy: 0.9927 - val_loss: 0.0835 - val_accuracy: 0.9879 Epoch 369/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0587 - accuracy: 0.9915 - val_loss: 0.0783 - val_accuracy: 0.9873 Epoch 370/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0782 - accuracy: 0.9912 - val_loss: 0.0966 - val_accuracy: 0.9873 Epoch 371/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0476 - accuracy: 0.9919 - val_loss: 0.0803 - val_accuracy: 0.9876 Epoch 372/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0485 - accuracy: 0.9925 - val_loss: 0.0910 - val_accuracy: 0.9877 Epoch 373/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0552 - accuracy: 0.9914 - val_loss: 0.1070 - val_accuracy: 0.9868 Epoch 374/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0481 - accuracy: 0.9919 - val_loss: 0.0915 - val_accuracy: 0.9882 Epoch 375/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0478 - accuracy: 0.9925 - val_loss: 0.0944 - val_accuracy: 0.9878 Epoch 376/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0483 - accuracy: 0.9921 - val_loss: 0.0863 - val_accuracy: 0.9877 Epoch 377/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0494 - accuracy: 0.9918 - val_loss: 0.0935 - val_accuracy: 0.9869 Epoch 378/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0578 - accuracy: 0.9912 - val_loss: 0.0911 - val_accuracy: 0.9873 Epoch 379/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0545 - accuracy: 0.9912 - val_loss: 0.0852 - val_accuracy: 0.9878 Epoch 380/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0424 - accuracy: 0.9933 - val_loss: 0.1014 - val_accuracy: 0.9874 Epoch 381/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0601 - accuracy: 0.9922 - val_loss: 0.0810 - val_accuracy: 0.9877 Epoch 382/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0573 - accuracy: 0.9913 - val_loss: 0.0818 - val_accuracy: 0.9874 Epoch 383/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0507 - accuracy: 0.9921 - val_loss: 0.1079 - val_accuracy: 0.9887 Epoch 384/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0575 - accuracy: 0.9917 - val_loss: 0.0927 - val_accuracy: 0.9880 Epoch 385/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0465 - accuracy: 0.9921 - val_loss: 0.0948 - val_accuracy: 0.9873 Epoch 386/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0519 - accuracy: 0.9921 - val_loss: 0.0893 - val_accuracy: 0.9876 Epoch 387/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0428 - accuracy: 0.9923 - val_loss: 0.0895 - val_accuracy: 0.9868 Epoch 388/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0574 - accuracy: 0.9916 - val_loss: 0.0854 - val_accuracy: 0.9875 Epoch 389/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0470 - accuracy: 0.9923 - val_loss: 0.0942 - val_accuracy: 0.9873 Epoch 390/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0555 - accuracy: 0.9917 - val_loss: 0.0895 - val_accuracy: 0.9876 Epoch 391/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0615 - accuracy: 0.9910 - val_loss: 0.1044 - val_accuracy: 0.9885 Epoch 392/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0777 - accuracy: 0.9905 - val_loss: 0.1180 - val_accuracy: 0.9867 Epoch 393/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0739 - accuracy: 0.9904 - val_loss: 0.1006 - val_accuracy: 0.9870 Epoch 394/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0829 - accuracy: 0.9906 - val_loss: 0.0889 - val_accuracy: 0.9878 Epoch 395/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0595 - accuracy: 0.9907 - val_loss: 0.0963 - val_accuracy: 0.9869 Epoch 396/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0674 - accuracy: 0.9910 - val_loss: 0.0971 - val_accuracy: 0.9866 Epoch 397/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0514 - accuracy: 0.9913 - val_loss: 0.0949 - val_accuracy: 0.9865 Epoch 398/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0438 - accuracy: 0.9922 - val_loss: 0.0820 - val_accuracy: 0.9873 Epoch 399/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0585 - accuracy: 0.9916 - val_loss: 0.0967 - val_accuracy: 0.9870 Epoch 400/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0573 - accuracy: 0.9918 - val_loss: 0.0956 - val_accuracy: 0.9860 Epoch 401/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0624 - accuracy: 0.9922 - val_loss: 0.0845 - val_accuracy: 0.9866 Epoch 402/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0531 - accuracy: 0.9916 - val_loss: 0.1288 - val_accuracy: 0.9879 Epoch 403/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0477 - accuracy: 0.9923 - val_loss: 0.0999 - val_accuracy: 0.9879 Epoch 404/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0467 - accuracy: 0.9919 - val_loss: 0.1138 - val_accuracy: 0.9866 Epoch 405/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0578 - accuracy: 0.9922 - val_loss: 0.0997 - val_accuracy: 0.9871 Epoch 406/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0683 - accuracy: 0.9901 - val_loss: 0.1156 - val_accuracy: 0.9871 Epoch 407/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0554 - accuracy: 0.9915 - val_loss: 0.1025 - val_accuracy: 0.9868 Epoch 408/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0483 - accuracy: 0.9917 - val_loss: 0.1084 - val_accuracy: 0.9875 Epoch 409/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0588 - accuracy: 0.9909 - val_loss: 0.1243 - val_accuracy: 0.9861 Epoch 410/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0655 - accuracy: 0.9905 - val_loss: 0.1162 - val_accuracy: 0.9871 Epoch 411/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0705 - accuracy: 0.9915 - val_loss: 0.1108 - val_accuracy: 0.9871 Epoch 412/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0762 - accuracy: 0.9909 - val_loss: 0.1126 - val_accuracy: 0.9870 Epoch 413/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0528 - accuracy: 0.9916 - val_loss: 0.1126 - val_accuracy: 0.9873 Epoch 414/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9918 - val_loss: 0.1133 - val_accuracy: 0.9876 Epoch 415/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0463 - accuracy: 0.9917 - val_loss: 0.1087 - val_accuracy: 0.9867 Epoch 416/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0546 - accuracy: 0.9915 - val_loss: 0.0921 - val_accuracy: 0.9872 Epoch 417/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0561 - accuracy: 0.9924 - val_loss: 0.1213 - val_accuracy: 0.9872 Epoch 418/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0523 - accuracy: 0.9917 - val_loss: 0.0921 - val_accuracy: 0.9860 Epoch 419/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0555 - accuracy: 0.9910 - val_loss: 0.0899 - val_accuracy: 0.9875 Epoch 420/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0585 - accuracy: 0.9919 - val_loss: 0.0960 - val_accuracy: 0.9871 Epoch 421/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0446 - accuracy: 0.9927 - val_loss: 0.1071 - val_accuracy: 0.9876 Epoch 422/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0785 - accuracy: 0.9922 - val_loss: 0.0868 - val_accuracy: 0.9875 Epoch 423/500 469/469 [==============================] - 2s 5ms/step - loss: 0.0498 - accuracy: 0.9915 - val_loss: 0.0905 - val_accuracy: 0.9885 Epoch 424/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0523 - accuracy: 0.9915 - val_loss: 0.0846 - val_accuracy: 0.9880 Epoch 425/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0494 - accuracy: 0.9928 - val_loss: 0.0909 - val_accuracy: 0.9876 Epoch 426/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0529 - accuracy: 0.9917 - val_loss: 0.1041 - val_accuracy: 0.9828 Epoch 427/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0516 - accuracy: 0.9922 - val_loss: 0.0982 - val_accuracy: 0.9880 Epoch 428/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0467 - accuracy: 0.9926 - val_loss: 0.0785 - val_accuracy: 0.9885 Epoch 429/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9916 - val_loss: 0.0950 - val_accuracy: 0.9881 Epoch 430/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0605 - accuracy: 0.9909 - val_loss: 0.1117 - val_accuracy: 0.9879 Epoch 431/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0600 - accuracy: 0.9911 - val_loss: 0.1007 - val_accuracy: 0.9876 Epoch 432/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0739 - accuracy: 0.9913 - val_loss: 0.1002 - val_accuracy: 0.9869 Epoch 433/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0487 - accuracy: 0.9920 - val_loss: 0.0899 - val_accuracy: 0.9871 Epoch 434/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0482 - accuracy: 0.9924 - val_loss: 0.1103 - val_accuracy: 0.9865 Epoch 435/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0493 - accuracy: 0.9927 - val_loss: 0.0979 - val_accuracy: 0.9870 Epoch 436/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0510 - accuracy: 0.9918 - val_loss: 0.1004 - val_accuracy: 0.9880 Epoch 437/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0451 - accuracy: 0.9921 - val_loss: 0.0955 - val_accuracy: 0.9867 Epoch 438/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0547 - accuracy: 0.9920 - val_loss: 0.1086 - val_accuracy: 0.9867 Epoch 439/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0482 - accuracy: 0.9917 - val_loss: 0.0912 - val_accuracy: 0.9871 Epoch 440/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0667 - accuracy: 0.9912 - val_loss: 0.1098 - val_accuracy: 0.9864 Epoch 441/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0506 - accuracy: 0.9921 - val_loss: 0.1144 - val_accuracy: 0.9850 Epoch 442/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9911 - val_loss: 0.1218 - val_accuracy: 0.9869 Epoch 443/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1562 - accuracy: 0.9910 - val_loss: 0.1236 - val_accuracy: 0.9875 Epoch 444/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0721 - accuracy: 0.9919 - val_loss: 0.1253 - val_accuracy: 0.9888 Epoch 445/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0603 - accuracy: 0.9912 - val_loss: 0.1260 - val_accuracy: 0.9865 Epoch 446/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0543 - accuracy: 0.9911 - val_loss: 0.1131 - val_accuracy: 0.9871 Epoch 447/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0560 - accuracy: 0.9914 - val_loss: 0.1055 - val_accuracy: 0.9875 Epoch 448/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0488 - accuracy: 0.9923 - val_loss: 0.1040 - val_accuracy: 0.9876 Epoch 449/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0517 - accuracy: 0.9919 - val_loss: 0.0916 - val_accuracy: 0.9876 Epoch 450/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1061 - accuracy: 0.9921 - val_loss: 0.0985 - val_accuracy: 0.9869 Epoch 451/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0498 - accuracy: 0.9920 - val_loss: 0.1043 - val_accuracy: 0.9875 Epoch 452/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0711 - accuracy: 0.9916 - val_loss: 0.1023 - val_accuracy: 0.9873 Epoch 453/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0615 - accuracy: 0.9916 - val_loss: 0.0927 - val_accuracy: 0.9881 Epoch 454/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0535 - accuracy: 0.9915 - val_loss: 0.0796 - val_accuracy: 0.9883 Epoch 455/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0468 - accuracy: 0.9919 - val_loss: 0.1241 - val_accuracy: 0.9885 Epoch 456/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0624 - accuracy: 0.9916 - val_loss: 0.0981 - val_accuracy: 0.9878 Epoch 457/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0538 - accuracy: 0.9919 - val_loss: 0.0998 - val_accuracy: 0.9884 Epoch 458/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0495 - accuracy: 0.9918 - val_loss: 0.1100 - val_accuracy: 0.9871 Epoch 459/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0537 - accuracy: 0.9921 - val_loss: 0.0930 - val_accuracy: 0.9879 Epoch 460/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0538 - accuracy: 0.9919 - val_loss: 0.0889 - val_accuracy: 0.9874 Epoch 461/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0491 - accuracy: 0.9922 - val_loss: 0.0981 - val_accuracy: 0.9877 Epoch 462/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0463 - accuracy: 0.9922 - val_loss: 0.1004 - val_accuracy: 0.9881 Epoch 463/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0462 - accuracy: 0.9923 - val_loss: 0.0992 - val_accuracy: 0.9854 Epoch 464/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0405 - accuracy: 0.9930 - val_loss: 0.1144 - val_accuracy: 0.9860 Epoch 465/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0723 - accuracy: 0.9928 - val_loss: 0.1198 - val_accuracy: 0.9869 Epoch 466/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0582 - accuracy: 0.9913 - val_loss: 0.1192 - val_accuracy: 0.9864 Epoch 467/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0435 - accuracy: 0.9920 - val_loss: 0.1355 - val_accuracy: 0.9872 Epoch 468/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0577 - accuracy: 0.9917 - val_loss: 0.1092 - val_accuracy: 0.9866 Epoch 469/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0520 - accuracy: 0.9920 - val_loss: 0.1024 - val_accuracy: 0.9875 Epoch 470/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0487 - accuracy: 0.9922 - val_loss: 0.0923 - val_accuracy: 0.9876 Epoch 471/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0424 - accuracy: 0.9926 - val_loss: 0.1076 - val_accuracy: 0.9869 Epoch 472/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0495 - accuracy: 0.9919 - val_loss: 0.1008 - val_accuracy: 0.9878 Epoch 473/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0442 - accuracy: 0.9927 - val_loss: 0.0977 - val_accuracy: 0.9881 Epoch 474/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0580 - accuracy: 0.9920 - val_loss: 0.1003 - val_accuracy: 0.9869 Epoch 475/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0489 - accuracy: 0.9924 - val_loss: 0.0990 - val_accuracy: 0.9874 Epoch 476/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0505 - accuracy: 0.9923 - val_loss: 0.1158 - val_accuracy: 0.9876 Epoch 477/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0513 - accuracy: 0.9921 - val_loss: 0.1089 - val_accuracy: 0.9877 Epoch 478/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0549 - accuracy: 0.9927 - val_loss: 0.1085 - val_accuracy: 0.9876 Epoch 479/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0542 - accuracy: 0.9922 - val_loss: 0.1029 - val_accuracy: 0.9868 Epoch 480/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0594 - accuracy: 0.9911 - val_loss: 0.0889 - val_accuracy: 0.9882 Epoch 481/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0571 - accuracy: 0.9915 - val_loss: 0.0940 - val_accuracy: 0.9874 Epoch 482/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0498 - accuracy: 0.9920 - val_loss: 0.0984 - val_accuracy: 0.9877 Epoch 483/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0649 - accuracy: 0.9920 - val_loss: 0.0939 - val_accuracy: 0.9880 Epoch 484/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0561 - accuracy: 0.9917 - val_loss: 0.0878 - val_accuracy: 0.9881 Epoch 485/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0497 - accuracy: 0.9922 - val_loss: 0.1071 - val_accuracy: 0.9881 Epoch 486/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0578 - accuracy: 0.9916 - val_loss: 0.1034 - val_accuracy: 0.9871 Epoch 487/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0499 - accuracy: 0.9921 - val_loss: 0.0936 - val_accuracy: 0.9875 Epoch 488/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0635 - accuracy: 0.9913 - val_loss: 0.1096 - val_accuracy: 0.9868 Epoch 489/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0483 - accuracy: 0.9924 - val_loss: 0.1026 - val_accuracy: 0.9887 Epoch 490/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0563 - accuracy: 0.9922 - val_loss: 0.1009 - val_accuracy: 0.9874 Epoch 491/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0544 - accuracy: 0.9923 - val_loss: 0.1077 - val_accuracy: 0.9883 Epoch 492/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0492 - accuracy: 0.9931 - val_loss: 0.1044 - val_accuracy: 0.9880 Epoch 493/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0557 - accuracy: 0.9921 - val_loss: 0.0881 - val_accuracy: 0.9882 Epoch 494/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0580 - accuracy: 0.9924 - val_loss: 0.0956 - val_accuracy: 0.9884 Epoch 495/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0553 - accuracy: 0.9919 - val_loss: 0.1043 - val_accuracy: 0.9877 Epoch 496/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0470 - accuracy: 0.9920 - val_loss: 0.0857 - val_accuracy: 0.9873 Epoch 497/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0428 - accuracy: 0.9928 - val_loss: 0.0771 - val_accuracy: 0.9879 Epoch 498/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0433 - accuracy: 0.9927 - val_loss: 0.0844 - val_accuracy: 0.9881 Epoch 499/500 469/469 [==============================] - 2s 4ms/step - loss: 0.0645 - accuracy: 0.9915 - val_loss: 0.0935 - val_accuracy: 0.9876 Epoch 500/500 469/469 [==============================] - 2s 4ms/step - loss: 0.1125 - accuracy: 0.9918 - val_loss: 0.0836 - val_accuracy: 0.9874
Visualize Dropout¶
In [ ]:
train_err = history.history['loss']
test_err = history.history['val_loss']
train_errd = historyd.history['loss']
test_errd = historyd.history['val_loss']
epochs = range(0,500)
plt.figure(figsize=(20,10))
plt.plot(epochs, train_err, 'b', label='No Dropout: Train Loss * 1000', linewidth=3)
plt.plot(epochs, test_err, 'b', label='No Dropout: Test Loss')
plt.plot(epochs, train_errd, 'r', label='Dropout: Train Loss * 1000', linewidth=3)
plt.plot(epochs, test_errd, 'r', label='Dropout: Test Loss')
plt.title('1024X5 Logistic')
plt.xlabel('Epoch')
plt.ylabel('Cross Entropy Error')
plt.legend()
plt.show()
In [ ]:
train_err = history1.history['loss']
test_err = history1.history['val_loss']
train_errd = historyd1.history['loss']
test_errd = historyd1.history['val_loss']
epochs = range(0,500)
plt.figure(figsize=(20,10))
plt.axis([None, None, 0, 1])
plt.plot(epochs, train_err, 'b', label='No Dropout: Train Loss * 1000', linewidth=3)
plt.plot(epochs, test_err, 'b', label='No Dropout: Test Loss')
plt.plot(epochs, train_errd, 'r', label='Dropout: Train Loss * 1000', linewidth=3)
plt.plot(epochs, test_errd, 'r', label='Dropout: Test Loss')
plt.title('1024X5 ReLU')
plt.xlabel('Epoch')
plt.ylabel('Cross Entropy Error')
plt.legend()
plt.show()
In [ ]: